CN112488413A - AWA-DRCN-based population spatialization method - Google Patents
AWA-DRCN-based population spatialization method Download PDFInfo
- Publication number
- CN112488413A CN112488413A CN202011460055.2A CN202011460055A CN112488413A CN 112488413 A CN112488413 A CN 112488413A CN 202011460055 A CN202011460055 A CN 202011460055A CN 112488413 A CN112488413 A CN 112488413A
- Authority
- CN
- China
- Prior art keywords
- population
- data
- model
- drcn
- awa
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 239000011159 matrix material Substances 0.000 claims abstract description 13
- 230000008569 process Effects 0.000 claims abstract description 9
- 238000007781 pre-processing Methods 0.000 claims abstract description 3
- 230000006870 function Effects 0.000 claims description 15
- 238000010276 construction Methods 0.000 claims description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 6
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 abstract description 3
- 238000013528 artificial neural network Methods 0.000 abstract 1
- 238000011160 research Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000007637 random forest analysis Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- DFPOZTRSOAQFIK-UHFFFAOYSA-N S,S-dimethyl-beta-propiothetin Chemical compound C[S+](C)CCC([O-])=O DFPOZTRSOAQFIK-UHFFFAOYSA-N 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 201000010099 disease Diseases 0.000 description 1
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Strategic Management (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- Tourism & Hospitality (AREA)
- Economics (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a population spatialization method based on an AWA-DRCN (active matrix architecture-DRCN), belonging to the technical field of Internet and computers. The method comprises the following steps: s1: acquiring a population data set and driving factor data, and preprocessing the population data set and the driving factor data; s2: the process of coarse removal: adopting an AWA model to realize global feature learning and finishing rasterization of district-county-level census data; s3: the process of extracting essence: and constructing a high-efficiency sub-pixel convolution neural network DRCN model, taking the rasterized population data and the driving factor data in the step S2 as characteristics, and inputting the processed gridded population raster data serving as a label into the DRCN model for mixed learning to realize global and local characteristics so as to obtain population spatialization results in spatial distribution. The AWA + DRCN model adopted by the invention has high simulation precision and the error between the space residual error values is minimum.
Description
Technical Field
The invention belongs to the technical field of internet and computers, and relates to a population spatialization method based on AWA-DRCN.
Background
Population spatialization is similar to the early downscaling processing process of climate data, and is a downscaling spatialization technology for converting a statistical population with low spatial resolution at a certain time point in a research area into a population distribution with high resolution close to the real population regional distribution. Because the spatial limitation problem of census data is solved to a certain extent by population spatialization data, the population spatialization data is widely applied to the aspects of disease disaster management, city planning and the like. Summarizing, the current population spatialization modeling technology is roughly divided into two main trends of global model construction and local model construction.
The global model construction process meets the requirement of low-resolution data by calculating the average population in the area, and the population statistical data is converted into a fine space unit by adopting a spatial interpolation method; with the development of remote sensing and GIS technology, multi-source heterogeneous data such as ETM images, elevations, gradients, night light data, Euclidean distances between land and rivers, land classification and the like are gradually applied, wherein the night light data such as DMSP-OLP, NPP/VIIRS and the difference between DMSP-OLP and NPP/VIIRS are frequently researched; and then combining the partitioned density mapping with a tree model and a multiple regression model. Random forest models are widely used for population spatialization research due to the advantages of high flexibility, scalability to variable importance degree and the like. And compared with a random forest, the multiple regression model considers the problem of correlation among different influence factors. In view of the problem that the global model is difficult to characterize the heterogeneity of population spatial distribution, local models such as geographical weighted regression and super-resolution convolutional network are introduced into population spatialization research practice. The geographical weighted regression model reduces the simulation precision due to the influence of complex terrain, and Zong Zefang et al introduces a super-resolution convolutional neural network model (SRCNN) into the astronomical population spatialization research of Shanghai city for the first time based on the commonality of single-image super-resolution and scale reduction technology, thereby obtaining better results. Thomas et al attempts to generate high resolution climate change predictions by single image super resolution by SRCNN stacking elevation. However, the SRCNN model has the problems of excessive parameters of a nonlinear mapping step and certain information loss caused by large filters, and local features can be better learned by improving a convolution structure or adopting sub-pixel convolution.
A single model can only extract global or local features independently, and cannot get rid of the influence of complex terrain, so that a method capable of realizing high-resolution population spatialization is urgently needed at present.
Disclosure of Invention
In view of the above, the present invention provides a population spatialization method based on AWA-DRCN, which is used for making up the defects of insufficient global feature learning and local feature generalization of a single model, so that a spatialization result is closer to tag data and closer to a population distribution reality.
In order to achieve the purpose, the invention provides the following technical scheme:
a population spatialization method based on AWA-ESPCN specifically comprises the following steps:
s1: acquiring a population data set and driving factor data, and preprocessing the population data set and the driving factor data;
s2: the process of coarse removal: adopting an Area weighted average rasterization (AWA) model to realize global feature learning and complete rasterization of district-level census data;
s3: the process of extracting essence: and (3) constructing a DRCN (deep recursive convolutional neural network) model, inputting the rasterized population data and 9 pieces of driving factor data in the step (S2) as characteristics, inputting the processed gridded population raster data serving as a label into the DRCN model for mixed learning to realize global and local characteristics, and obtaining population spatialization results on spatial distribution.
Further, in step S1, the demographic data set includes demographic data and high-resolution grid demographic data.
Further, in step S1, the driving factor data includes a natural factor, a social factor, and a distance factor.
Further, the natural factors include: elevation, gradient and land utilization types, wherein the land utilization types are divided into water, rivers, cultivated land and forest land; the social factors include: night lights and residential points; the distance factor includes: european distance between land and river.
Further, in step S2, the expression of the AWA model used is:
among them, LRkiRepresenting the population number, P, of the ith grid in the kth administrationkRepresents the kth administrative Unit population, SkDenotes the area of the kth administration unit (unit: hm)2)。
Further, constructing a DRCN model specifically comprises:
(1) the network takes the interpolated input image (to the required size) as input x and predicts the target image y; the object of the invention is to learn a model f with a prediction value Y ═ f (x), where Y is its prediction estimate for the target image Y; let f1、f2、f3Respectively, the subnet functions are represented: embedding, reasoning and reconstructing; the model constructed consists of three functions: f (x) ═ f3(f2(f1(x) ); embedded net f1(x) Taking an input vector x and calculating a matrix output H0It is a reasoning network f2The input of (1); h for hidden layer value-1Represents; the net-nesting formula is as follows:
H-1＝max(0,W-1*x+b-1)
H0＝max(0,W0*H-1+b0)
f1(x)＝H0
where operator denotes convolution, max (0,) corresponds to ReLU; the weight and bias matrix is W-1、W0And b-1、b0；
(2) Inference network f2Get the input matrix H0And calculates the matrix output HD. Here we use the same weight and bias matrices W and b for all operations. Let g denote the function of a single recursive modeling of the recursive layer: g (H) ═ max (0, W × H + b); the recurrence relation is
Hd＝g(Hd-1)＝max(0,W*Hd-1+b)
For D12Composition corresponding to function g: where the operator o represents a combination of functions, gdRepresents the product of g and d;
f2(H)＝(gogo…o)g(H)＝gD(H)
(3) reconstruction net f3Get the hidden state of input HDAnd outputs a target image (high resolution). Roughly speaking, a reconstruction network is the inverse operation of an embedded network, and the formula is as follows:
HD+1＝max(0,WD+1*HD+bD+1)
Y＝max(0,WD+2*HD+1+bD+2)
f3(H)＝Y
taking into account the influence of the influencing factors on the population distribution, the experiment takes the grid population and the different influencing factors as inputs to the DRCN in the form of channels, the number of which depends on the amount of helper data. Primary output of the model is obtained through three parts, namely an embedded network, an inference network and a reconstructed network. The task of population spatialization is to make the model output and the actual population data as similar as possible, so that the mean square error of the model output and the actual population distribution is minimized by adopting a standard back propagation random gradient descent method, and a high-resolution population distribution result is obtained.
The invention has the beneficial effects that:
(1) under the combined action of the natural environment factor and the social and economic factor, the AWA + DRCN model coupling adopted by the invention makes up the defects of insufficient global feature learning, generalization of local features and the like of a single model to a certain extent.
(2) The simulation precision of the AWA + DRCN coupling model adopted by the invention is highest, and the error of the space residual error value interval is smallest.
(3) The area weighted average rasterized population adopted by the invention well learns the global population characteristics (aiming at the area with small population density), and the driving factor as auxiliary data can improve the precision of the training model and reduce the residual error.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention. The objectives and other advantages of the invention may be realized and attained by the means of the instrumentalities and combinations particularly pointed out hereinafter.
Drawings
For the purposes of promoting a better understanding of the objects, aspects and advantages of the invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings in which:
FIG. 1 is a flow chart of the AWA-DRCN-based population spatialization method of the present invention;
FIG. 2 is a schematic view of the location of an investigation region;
fig. 3 is a graph of drive factor data.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention in a schematic way, and the features in the following embodiments and examples may be combined with each other without conflict.
Referring to fig. 1 to 3, the present embodiment designs a population spatialization method, which specifically includes the following steps:
s1: the population data set and the driver factor data are acquired and preprocessed.
The regional adaptability of the spatialization model can be better verified by selecting Chongqing city as a research region. The demographic data set used by the present embodiment includes demographic data and high-resolution grid demographic data. The demographic data middle area and county household population data are sourced from http:// www.nature.com/sdata, and the high-resolution grid population is sourced from world population project (WorldPop) https:// www.worldpop.org as model tag data.
The driving factor data used in this embodiment includes 9 driving factor data including natural factors (elevation, gradient, land utilization type), social factors (night light, residential point) and distance factors (european distance between land and river), where the land utilization type in the natural factors is divided into 4 categories (water body, river, farmland, forest land).
Elevation, grade, land use type data are derived from global 30 meter resolution surface coverage data (GlobeLand 30); the night light data is a 100m resolution calibration value (0-6300) obtained by multiplying standard uncalibrated DMSP light source data (0-63) by 100, and the residential point data is a 100m resolution urban and rural residential point generated by a student of Jermeniah J and the like by using a random forest model; vector data such as rivers, administrative boundaries and the like are derived from free street view data, so that the Euclidean distance between the land and the rivers can be calculated; the land types (7 types of cultivated land, woodland, shrub woodland, grassland, wetland, water body and construction land) are combined into four main types, namely the wetland and the water body are used as an unmanned water area, the forest and the shrub woodland are combined into a woodland, the cultivated land and the grassland are combined into the cultivated land, the construction land is kept unchanged, and simultaneously, each land type is extracted to form a graph 3.
S2: the process of coarse removal: and (3) realizing global feature learning by adopting an Area weighted average rasterization (AWA) model, and completing 500m resolution rasterization of the county-level census data. In this embodiment, the result of "bolding" is recorded as a Low Resolution population (LR population).
Area Weighted Average (AWA) method means dividing the total population number of a statistical unit by the total Area of the statistical unit (statistical unit Area unit: hm)2). The statistical population is uniformly distributed in administrative units of county level by AWA method (the unit of population density is: person/hm)2). The calculation formula is as follows
Among them, LRkiRepresenting the population number, P, of the ith grid in the kth administrationkRepresents the kth administrative Unit population, SkDenotes the area of the kth administration unit (unit: hm)2)。
Constructing a DRCN model, which specifically comprises the following steps:
(1) the network takes the interpolated input image (to the required size) as input x and predicts the target image y; the object of the invention is to learn a model f with a prediction value Y ═ f (x), where Y is its prediction estimate for the target image Y; let f1、f2、f3Respectively, the subnet functions are represented: embedding, reasoning and reconstructing; the model constructed consists of three functions: f (x) ═ f3(f2(f1(x) ); embedded net f1(x) Taking an input vector x and calculating a matrix output H0It is a reasoning network f2The input of (1); h for hidden layer value-1Represents; the net-nesting formula is as follows:
H-1＝max(0,W-1*x+b-1)
H0＝max(0,W0*H-1+b0)
f1(x)＝H0
where operator denotes convolution, max (0,) corresponds to ReLU; the weight and bias matrix is W-1、W0And b-1、b0；
(2) Inference network f2Get the input matrix H0And calculates the matrix output HD. Here we use the same weight and bias matrices W and b for all operations. Let g denote the function of a single recursive modeling of the recursive layer: g (H) ═ max (0, W × H + b); the recurrence relation is
Hd＝g(Hd-1)＝max(0,W*Hd-1+b)
For D12Composition corresponding to function g: where the operator o represents a combination of functions, gdRepresents the product of g and d;
f2(H)＝(gogo…o)g(H)＝gD(H)
(3) reconstruction net f3Get the hidden state of input HDAnd outputs a target image (high resolution). Roughly speaking, a reconstruction network is the inverse of an embedded network, and is formulated asThe following:
HD+1＝max(0,WD+1*HD+bD+1)
Y＝max(0,WD+2*HD+1+bD+2)
f3(H)＝Y
taking into account the influence of the influencing factors on the population distribution, the experiment takes the grid population and the different influencing factors as inputs to the DRCN in the form of channels, the number of which depends on the amount of helper data. Primary output of the model is obtained through three parts, namely an embedded network, an inference network and a reconstructed network. The task of population spatialization is to make the model output and the actual population data as similar as possible, so that the mean square error of the model output and the actual population distribution is minimized by adopting a standard back propagation random gradient descent method, and a high-resolution population distribution result is obtained.
Finally, the above embodiments are only intended to illustrate the technical solutions of the present invention and not to limit the present invention, and although the present invention has been described in detail with reference to the preferred embodiments, it will be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions, and all of them should be covered by the claims of the present invention.
Claims (6)
1. A AWA-DRCN-based population spatialization method is characterized by comprising the following steps:
s1: acquiring a population data set and driving factor data, and preprocessing the population data set and the driving factor data;
s2: the process of coarse removal: adopting an Area weighted average rasterization (AWA) model to realize global feature learning and complete rasterization of district-level census data;
s3: the process of extracting essence: and (4) constructing a DRCN model, taking the rasterized population data and the driving factor data in the step (S2) as characteristics, inputting the processed gridded population raster data serving as a label into the DRCN model for mixed learning to realize global and local characteristics, and obtaining population spatialization results on spatial distribution.
2. The population spatialization method according to claim 1, wherein in step S1, the population data set includes demographic data and high-resolution grid demographic data.
3. The population spatialization method according to claim 1, wherein in step S1, the driving factor data includes a natural factor, a social factor and a distance factor.
4. The population spatialization method according to claim 3, wherein the natural factors comprise: elevation, gradient and land utilization types, wherein the land utilization types are divided into water, rivers, cultivated land and forest land; the social factors include: night lights and residential points; the distance factor includes: european distance between land and river.
5. The population spatialization method according to claim 1, wherein in step S2, the expression of the AWA model is:
among them, LRkiRepresenting the population number, P, of the ith grid in the kth administrationkRepresents the kth administrative Unit population, SkThe area of the kth administration unit is shown.
6. The population spatialization method according to claim 1, wherein in step S3, constructing the DRCN model specifically includes:
(1) the network takes the interpolated input image as input x and predicts a target image y; the target is to learn a model f with a predicted value Y ═ f (x), where Y is its predicted estimate of the target image Y; let f1、f2、f3Respectively, the subnet functions are represented: embedding, reasoning and reconstructing; construction ofThe model of (a) consists of three functions: f (x) ═ f3(f2(f1(x) ); embedded net f1(x) Taking an input vector x and calculating a matrix output H0It is a reasoning network f2The input of (1); h for hidden layer value-1Represents; the net-nesting formula is as follows:
H-1＝max(0,W-1*x+b-1)
H0＝max(0,W0*H-1+b0)
f1(x)＝H0
where operator denotes convolution, max (0,) corresponds to ReLU; the weight and bias matrix is W-1、W0And b-1、b0；
(2) Inference network f2Get the input matrix H0And calculates the matrix output HD(ii) a The same weight and bias matrices W and b are used for all operations; let g denote the function of a single recursive modeling of the recursive layer: g (H) ═ max (0, W × H + b); the recurrence relation is
Hd＝g(Hd-1)＝max(0,W*Hd-1+b)
For D12Composition corresponding to function g: where the operator o represents a combination of functions, gdRepresents the product of g and d;
f2(H)＝(gogo…o)g(H)＝gD(H)
(3) reconstruction net f3Get the hidden state of input HDOutputting a target image; that is, the reconstruction network is the inverse operation of the embedded network, and the formula is as follows:
HD+1＝max(0,WD+1*HD+bD+1)
Y＝max(0,WD+2*HD+1+bD+2)
f3(H)＝Y
obtaining primary output of the model through an embedded network, an inference network and a reconstruction network; and minimizing the mean square error of the model output and the actual population distribution by adopting a standard back propagation random gradient descent method so as to obtain a high-resolution population distribution result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011460055.2A CN112488413A (en) | 2020-12-11 | 2020-12-11 | AWA-DRCN-based population spatialization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011460055.2A CN112488413A (en) | 2020-12-11 | 2020-12-11 | AWA-DRCN-based population spatialization method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112488413A true CN112488413A (en) | 2021-03-12 |
Family
ID=74916174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011460055.2A Pending CN112488413A (en) | 2020-12-11 | 2020-12-11 | AWA-DRCN-based population spatialization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112488413A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113344811A (en) * | 2021-05-31 | 2021-09-03 | 西南大学 | Multilayer convolution sparse coding weighted recursive denoising deep neural network and method |
| CN113554356A (en) * | 2021-09-18 | 2021-10-26 | 中国地质大学（武汉） | Population distribution estimation method and device based on multi-model fusion and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650618A (en) * | 2016-11-15 | 2017-05-10 | 中山大学 | Random forest model-based population data spatialization method |
| CN109376969A (en) * | 2018-12-14 | 2019-02-22 | 中南大学 | City fining population distribution dynamic prediction method and device based on deep learning |
| CN110458333A (en) * | 2019-07-18 | 2019-11-15 | 华南农业大学 | A kind of population spatial distribution prediction technique and system based on POIs data |
| CN110716998A (en) * | 2019-07-25 | 2020-01-21 | 首都师范大学 | Method for spatializing fine-scale population data |
-
2020
- 2020-12-11 CN CN202011460055.2A patent/CN112488413A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106650618A (en) * | 2016-11-15 | 2017-05-10 | 中山大学 | Random forest model-based population data spatialization method |
| CN109376969A (en) * | 2018-12-14 | 2019-02-22 | 中南大学 | City fining population distribution dynamic prediction method and device based on deep learning |
| CN110458333A (en) * | 2019-07-18 | 2019-11-15 | 华南农业大学 | A kind of population spatial distribution prediction technique and system based on POIs data |
| CN110716998A (en) * | 2019-07-25 | 2020-01-21 | 首都师范大学 | Method for spatializing fine-scale population data |
Non-Patent Citations (3)
| Title |
|---|
| YUTING LI等: "Research on Population Spatialization Method Based on PMST-SRCNN", 《E3S WEB OF CONFERENCES》 * |
| 孙强等: "低质量无约束人脸图像下的超分辨率摆正", 《计算机应用》 * |
| 游鹏: "基于AWA-ESPCN的人口空间化及模拟预测研究", 《中国优秀博硕士学位论文全文数据库(硕士) 社会科学Ⅱ辑》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113344811A (en) * | 2021-05-31 | 2021-09-03 | 西南大学 | Multilayer convolution sparse coding weighted recursive denoising deep neural network and method |
| CN113554356A (en) * | 2021-09-18 | 2021-10-26 | 中国地质大学（武汉） | Population distribution estimation method and device based on multi-model fusion and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Demiray et al. | D-SRGAN: DEM super-resolution with generative adversarial networks | |
| CN110263111B (en) | Priori knowledge-based space-time monitoring method for land utilization/coverage information | |
| CN106780089B (en) | Permanent basic farmland planning method based on neural network cellular automaton model | |
| CN111259898A (en) | Crop segmentation method based on unmanned aerial vehicle aerial image | |
| CN106716479A (en) | A computer implemented method of deriving performance from a local model | |
| CN114331842B (en) | DEM super-resolution reconstruction method combining topographic features | |
| CN112488413A (en) | AWA-DRCN-based population spatialization method | |
| CN113902580B (en) | Historical farmland distribution reconstruction method based on random forest model | |
| Niese et al. | Procedural urban forestry | |
| CN116090065A (en) | Digital twinning-based smart city greening design method and device | |
| Tezel et al. | Accurate assessment of protected area boundaries for land use planning using 3D GIS | |
| Zhang et al. | SolarGAN: Synthetic annual solar irradiance time series on urban building facades via Deep Generative Networks | |
| Zhao et al. | Examining land-use change trends in yucheng district, Ya’an city, China, using ANN-CA modeling | |
| Crols et al. | Downdating high-resolution population density maps using sealed surface cover time series | |
| Mansor et al. | Optimization of land use suitability for agriculture using integrated geospatial model and genetic algorithms | |
| CN102880753A (en) | Method for converting land utilization spatial characteristic scale based on fractal dimension | |
| CN114418243B (en) | Distributed new energy cloud grid prediction method and system | |
| CN115952743A (en) | Multi-source precipitation data collaborative downscaling method and system coupled with random forest and HASM | |
| CN115049160A (en) | Method and system for estimating carbon emission of plain industrial city by using space-time big data | |
| CN114324410A (en) | Multi-terrain microwave remote sensing soil humidity downscaling method | |
| Handayani et al. | Modelling of land use change in indramayu district, west Java Province | |
| CN114360739B (en) | Dengue risk prediction method based on remote sensing cloud computing and deep learning | |
| Rothschedl et al. | Calculating Shadows with U-Nets for Urban Environments (Short Paper) | |
| CN116152503B (en) | Urban sky visible field online extraction method and system facing streetscape | |
| CN117593470B (en) | Street view reconstruction method and system based on AI model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210312 |