CN114662774A - City block vitality prediction method, storage medium and terminal - Google Patents
City block vitality prediction method, storage medium and terminal Download PDFInfo
- Publication number
- CN114662774A CN114662774A CN202210351848.3A CN202210351848A CN114662774A CN 114662774 A CN114662774 A CN 114662774A CN 202210351848 A CN202210351848 A CN 202210351848A CN 114662774 A CN114662774 A CN 114662774A
- Authority
- CN
- China
- Prior art keywords
- block
- data
- vitality
- city
- value
- 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
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000003860 storage Methods 0.000 title claims abstract description 10
- 238000012549 training Methods 0.000 claims abstract description 42
- 230000000694 effects Effects 0.000 claims abstract description 33
- 238000010586 diagram Methods 0.000 claims description 25
- 238000005520 cutting process Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 4
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000010438 heat treatment Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 238000010187 selection method Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 238000011835 investigation Methods 0.000 claims 1
- 238000010276 construction Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 18
- 238000009826 distribution Methods 0.000 description 9
- 238000012360 testing method Methods 0.000 description 7
- 238000012417 linear regression Methods 0.000 description 5
- 238000011160 research Methods 0.000 description 5
- 230000001419 dependent effect Effects 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 239000003795 chemical substances by application Substances 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000009966 trimming 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
- G06N20/00—Machine learning
-
- 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)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- Development Economics (AREA)
- Software Systems (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Primary Health Care (AREA)
- Educational Administration (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a city block vitality prediction method, a storage medium and a terminal, and belongs to the technical field of city planning. A city block vitality prediction method comprises the following steps: acquiring a vector boundary of a block in a region to be detected from road network data, and taking the region in the boundary as a basic unit; acquiring built environment data of a block in a region to be detected from a basic unit as an independent variable; acquiring crowd activity intensity data of a block in a region to be detected, and quantifying the crowd activity intensity data to acquire an urban activity value; matching the built environment data of the basic units and the blocks with the crowd activity intensity data to construct a training database; training the training database by adopting a GBDT algorithm to obtain a city block vitality prediction model; and predicting the block vitality of the area to be detected based on the city block vitality prediction model. According to the method, the city vitality after future construction is predicted more accurately through the construction environment indexes of the blocks, so that a basis is provided for a planning and designing strategy.
Description
Technical Field
The invention relates to the technical field of urban planning, in particular to a city block vitality prediction method, a storage medium and a terminal.
Background
Urban vitality is the ability of the living environment to stimulate human-to-human interaction and is the most essential element for realizing urban quality of life. Thus, since the 50 s of the 20 th century, researchers and practitioners have explored urban vitality as an element of urban development and planning. The prediction of city vitality has important significance for the development and construction of cities, and can help city planners and managers to quantitatively evaluate the design strategy and the development mode of a planning area or an updating area. However, most research and application focuses on measuring urban vitality and the relationship between urban vitality and elements such as space, society and the like by using multi-source data, and related technical methods for urban vitality prediction are rarely involved. Therefore, a city block vitality prediction method is provided.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a city block vitality prediction method.
The purpose of the invention can be realized by the following technical scheme:
a city block vitality prediction method comprises the following steps:
acquiring a vector boundary of a block in a region to be detected from road network data, and taking the region in the vector boundary as a basic unit;
acquiring built environment data of a block in a region to be detected from a basic unit as an independent variable;
acquiring crowd activity intensity data of a block in a region to be detected, and quantifying the crowd activity intensity data to acquire a city activity value;
matching the built environment data of the basic unit and the blocks in the area to be detected with the crowd activity intensity data to construct a training database; wherein the input sample is an independent variable, and the output sample is an urban vitality value;
training the training database by adopting a GBDT regression algorithm to obtain a city block vitality prediction model;
and predicting the block vitality of the area to be detected based on the city block vitality prediction model.
Further, the crowd activity intensity data includes one or more of thermodynamic data, night light data, or social software check-in data.
Further, the step of quantifying the crowd activity intensity data to obtain the city activity value comprises the following steps:
acquiring remote sensing image map data and thermodynamic diagram data of a region to be detected; carrying out geographic registration on the acquired thermodynamic diagram and the remote sensing image map data by taking the remote sensing image map data as a reference; converting the pixel value of the registered thermodynamic diagram fourth wave band data into an integer, and then converting the thermodynamic diagram fourth wave band data into vector data; finally, summarizing and counting the fourth band value of each block in the region to be measured to obtain the block heat force value HiStreet heating power value HiNamely the city vitality value.
Further, the step of quantifying the crowd activity intensity data to obtain the city activity value comprises the following steps:
acquiring a night light image map of the area to be detected, and performing grid cutting on the night light image map according to the vector boundary of the area to be detected; converting the pixel value of the first wave band of the clipped night light image map into an integer, and then converting the first wave band data of the night light image map into vector data; finally, summarizing and counting the first waveband numerical value of the night light image map of each block in the region to be detected to obtain the block brightness value NiLuminance value N of blockiNamely the city vitality value.
Further, the step of quantifying the crowd activity intensity data to obtain the city activity value comprises the following steps:
acquiring social software sign-in data of an area to be detected, and screening out repeated data with the same ID, the same time and the same position from the social software sign-in data; counting check-in of each block in the region to be measured in the screened social software check-in dataNumber of times FiNumber of check-in times FiNamely the city vitality value.
Further, the training of the urban block vitality prediction model by adopting the GBDT regression algorithm for the training database comprises the following steps:
input training data set T { (x)1,y1,),(x2,y1,),...，(xN,yN,)},xN＝(xn,1,xn,2,...xn,k) (ii) a Where N represents the number of blocks in the area under study, xNThe built environment value representing the Nth block in the region to be measured in the training data set has xn,1,xn,2,.. et al total k independent variables, yNRepresenting the city vitality value of the Nth block in the region to be tested in the training data set;
selecting a loss function L, initializing a weak learner f0(x) Is a constant c, which should minimize the loss function L:
in the formula YiThe city vitality value of the ith block in the area to be measured;
sequentially training 1,2, the m weak learners, and solving a negative gradient value of a sample at the moment when the m weak learner is trained:
fitting the negative gradient value to generate a new weak learner, and obtaining a leaf node region R of the weak learnermjJ1, 2.. J for J1, 2.. J, the calculation:
updating the model to obtain fm(x)：
And (3) stopping training until the Mth weak learner is trained, and finally obtaining a strong learner gradient promotion decision tree, namely a city block vitality prediction model:
further, the independent variables include one or more of building density, volume fraction, greenfield fraction, function density, function mix, right diversity, distance to nearest bus stop, distance to nearest subway stop, distance to city center, street area, street compactness, average building footprint.
Further, the training database is optimized by adopting characteristic engineering operation, and the method comprises the following steps:
preprocessing data in a training database, and performing feature conversion on preprocessed built environment data;
scoring each feature according to the correlation by using a variance selection method in a filtering type, and setting a threshold value to select the feature;
and constructing the features by adopting a cross feature method, and verifying whether the selected features can improve the accuracy of prediction.
In another aspect, the present invention further provides a storage medium, wherein a plurality of programs are stored, and the programs are loaded and executed by a processor to implement any one of the above city block vitality prediction methods.
In a third aspect, the present invention further provides a city block vitality prediction terminal, including a processor, adapted to execute various programs, where the programs are loaded and executed by the processor to implement any one of the above city block vitality prediction methods.
The beneficial results of the invention are:
the method for predicting the urban block vitality accurately by adopting the machine learning method based on multi-source big data solves the problem of inaccurate prediction caused by the lack of urban vitality prediction or the adoption of a simple linear model in the prior art, and is specifically embodied in the following two aspects: the urban block vitality is predicted based on multi-source urban data, the coverage range is large, and the traditional statistical mode which wastes time and labor can be replaced; the strong learner is established by adopting a machine learning method to predict the urban vitality, the accuracy is higher, and the limitation and the error caused by a linear model can be avoided.
Drawings
The invention will be further described with reference to the accompanying drawings.
FIG. 1 is a schematic flow diagram of a method of the present application;
FIG. 2 is a distribution diagram of city vitality values in a neighborhood of an embodiment;
FIG. 3 is a graph of a street volume rate distribution for an embodiment;
FIG. 4 is a block density distribution diagram of an embodiment;
FIG. 5 is a graph of the average number of building floors for a block of an exemplary embodiment;
FIG. 6 is a graph of a block greenfield rate distribution for an embodiment;
FIG. 7 is a graph of a block green visibility graph according to an embodiment;
FIG. 8 is a plot of the sky openness in a neighborhood for an embodiment;
FIG. 9 is a graph of street function mix distribution for an embodiment;
FIG. 10 is a graph of an exemplary block functional density distribution;
FIG. 11 is a graph of a distribution of road network density for an embodiment case block;
FIG. 12 is a graph of a distribution of distance between a neighborhood and a city center for an embodiment;
fig. 13 is an embodiment block distance distribution diagram to nearest subway stations.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the description of the present invention, it is to be understood that the terms "opening," "upper," "lower," "thickness," "top," "middle," "length," "inner," "peripheral," and the like are used in an orientation or positional relationship that is merely for convenience in describing and simplifying the description, and do not indicate or imply that the referenced component or element must have a particular orientation, be constructed and operated in a particular orientation, and thus should not be considered as limiting the present invention.
In the description herein, references to the description of "one embodiment," "an example," "a specific example" or the like are intended to mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
The invention provides a city block vitality prediction method, which comprises the following steps:
1. road network data can be acquired through ways such as an OSM, a DIVA-GIS, a road network database of each province, a design unit, a transportation department and the like; taking an OSM traffic network as an example, dividing an area to be detected into blocks according to the OSM traffic network, and the method specifically comprises the following steps:
screening out expressways, main roads, primary roads, secondary roads and tertiary roads in a road network, carrying out topology processing on the road network and trimming suspended roads and independent road sections in the road network;
dividing roads into three levels according to the actual conditions of the areas, wherein the expressway and the main road are in a first level, the primary road and the secondary road are in a second level, and the tertiary road is in a third level, and respectively establishing buffer areas of 40m, 20m and 10m for the three levels of roads so as to generate a road space;
removing the road space, generating an independent vector boundary, and taking an area in the vector boundary as a basic unit;
2. road network data are processed through ArcGIS, GeoDa, uDig, OpenJump, QGIS and other software; ArcGIS software is selected to process the road network data; firstly, the distance from the block center of mass to the nearest subway station, the distance from the block center to the city center, the block area and the block compactness are calculated by utilizing the vector data of the block unit.
(1) The method for calculating the block compactness specifically comprises the following steps: and generating a circumscribed circle of the block according to the vector boundary of the block, and then dividing the block area by the circumscribed circle area of the block to obtain the block compactness.
(2) And acquiring building vector data for calculating the building density, the volume ratio and the average building base area of each block.
(3) And obtaining POI interest point data for calculating the function density and the function mixing degree of each block:
wherein P isiIs the proportion of the ith function type, namely dividing POI data into five types of residence, public service, business, office and traffic, calculating the proportion of each interest point quantity in the block to the total interest point quantity to obtain Phousing、Ppublic、Pcommercial、PofficeAnd Ptraffic。
(4) Obtaining population grid data to calculate population density, comprising the steps of:
acquiring population grid data, and performing grid cutting on the population grid data according to the vector boundary of the research range;
and finally, connecting the obtained vector surface file to the block units through a space connection tool, summarizing to obtain the block population quantity, and dividing the population quantity by the block area to obtain the population density.
(5) Acquiring the current land data to calculate the land diversity: and counting the number of different types of land in the block unit according to the current land map so as to measure the diversity of the land.
3. Acquiring crowd activity intensity data as a dependent variable, and quantifying to obtain an urban activity value ViThe crowd vitality data comprises thermodynamic diagram data and social software check-in dataThree data of night light data are taken as examples for explanation, and the specific steps comprise:
(1) street heating value HiObtaining:
acquiring remote sensing image map data and thermodynamic diagram data of a region to be detected; carrying out geographic registration on the acquired thermodynamic diagram and the remote sensing image map data by taking the remote sensing image map data as a reference; converting pixel values of the registered thermodynamic diagram fourth Band4 data into integer, and then converting the thermodynamic diagram fourth Band4 data into vector data; finally, summarizing and counting the fourth Band4 value of each block in the region to be measured to obtain the block heat value Hi(ii) a The data of the fourth Band4 of the thermodynamic diagram represents the heat of the thermodynamic diagram, namely the relative number and concentration degree of people.
(2) Street night brightness value NiObtaining:
acquiring a night light image map of the area to be detected, and performing grid cutting on the night light image map according to the vector boundary of the area to be detected; converting the pixel value of the cut night light image map first waveband Band1 into integer, and then converting the night light image map first waveband Band1 data into vector data; finally, summarizing and counting the first Band1 value of each block night light image in the region to be detected to obtain the block brightness value Ni；
The data of the first waveband Band1 of the night light image represents the brightness value of the night light image, and brighter places represent high regional light values and frequent human activities.
(3) Number of sign-ins FiObtaining:
obtaining social software check-in data of a region to be tested, and screening out repeated data with the same ID, the same time and the same position from the social software check-in data; counting the check-in times F of each block in the region to be detected from the screened social software check-in datai。
(4) Weighting one or more types of thermodynamic diagram data, social media sign-in data and night light data according to an entropy weight method, and then calculating to obtain an urban vitality value Vi：
4. The method comprises the following steps of registering and checking coordinates of built environment data, and building a training database, and specifically comprises the following steps: checking a coordinate system of the data, and unifying data coordinates of the coordinate system into a national geodetic coordinate system CGCS 2000;
and converting the raster data in the built-up environment data into vector data.
And connecting the constructed environment data into the basic unit, and classifying and summarizing to obtain the value of each index of the constructed environment.
5. The characteristic engineering operation performed on the training database specifically comprises the following steps:
data preprocessing: cleaning the constructed environment index data in the training database, mainly solving the problems of missing values, abnormal values, error values, sampling degree and the like;
scoring each feature according to the correlation by using a variance selection method in a filtering Filter, and setting a threshold value to select the feature;
and constructing the features by adopting a cross feature method, and verifying whether all the features can improve the accuracy of the pre-measurement.
6. The GBDT model is trained by utilizing the training database after the characteristic engineering operation, and the specific steps comprise:
inputting a training data set, selecting a suitable loss function L, e.g. a squared loss function, initializing a weak learner f0(x)：T＝{(X1,Y1,),(X2,Y1,),...,(XN,YN,)}
XN＝(Xn,1,Xn,2,...Xn,k)
L(y,f(x))
Where N represents the number of blocks in the area under study, XNThe independent variable representing each block in the training data set is the constructed environmental data, and has Xn,1,Xn,2,.. et al total k independent variables, YNRepresenting a dependent variable, i.e. a city activity value, for each block in the training dataset, wherein the constant c is a function of the initial lossL is the smallest value;
sequentially training 1,2., m weak learners, and when the m weak learners are trained, obtaining a negative gradient value of each sample i ═ 1,2., N at the time:
taking the obtained negative gradient value as a new true value of the sample, and taking the data
As training data of the weak learner, obtaining a new weak learner with a corresponding leaf node region of RmjJ1, 2, J, where J is the number of leaf nodes of the weak learner.
For leaf region J ═ 1,2.. J, calculate:
updating the model to obtain fm(x)：
Stopping training until the Mth weak learner is reached, and finally obtaining a strong learner gradient lifting decision tree:
bringing test data into fM(x) Finally obtaining the activity predicted value V of the test blockpi。
The first embodiment is as follows: the method is practiced by taking an old city area of a certain city as a research object, the total area is about 43.3 square kilometers, the district comprises a central area of the city and a plurality of commercial and cultural facilities, the crowd activity is rich, and the method is a better choice for predicting the activity of the city.
As shown in fig. 1, the embodiment provides a method for predicting the activity of a city block, which specifically includes the following steps:
1. the method comprises the steps of obtaining OSM traffic network data, dividing a research area into a plurality of blocks, taking the area in the block boundary as a basic unit, and enabling the number of the basic units of the implementation case to be 151.
2. Acquiring hundredth thermodynamic diagram data, wherein the hundredth thermodynamic diagram data acquired in the embodiment is from average raster data of 10 months, 21 days and one day in 2021, carrying out coordinate registration on the average raster data, converting pixel values of the registered thermodynamic diagram fourth Band4 data into integer types, and then converting the thermodynamic diagram fourth Band4 data into vector data; finally, summarizing and counting the numerical values of the fourth Band4 of each block in the region to be measured to obtain the block heat value Hi。
3. Acquiring night light data, wherein the night light data acquired in the embodiment is 130m x 130m raster data of Lojia I, converting the pixel value of the first waveband Band1 after being cut into integer, and converting the night light image first waveband Band1 data into vector data; finally, summarizing and counting the first Band1 value of each block night light image in the region to be detected to obtain the block brightness value Ni。
4. Obtaining green wave microblog sign-in data, wherein the green wave microblog sign-in data obtained in the embodiment are from punctiform vector data of 10 months in 2021, and repeating data with the same ID, the same time and the same position are screened out; then, the check-in times F of each basic unit are counted from the screened social software check-in datai。
5. H is to bei,Ni,FiAfter normalization, weights are calculated according to an entropy weight method and respectively are as follows: 0.419; 0.272; 0.309, and; then H is introducedi,Ni,FiRespectively multiplied by the weights to obtain the city vitality value ViAs shown in fig. 2.
6. Building vector data, land data, street view image data and POI interest point data are obtained, and a built environment index is calculated;
(1) calculating the volume ratio, the building density and the average building layer number by using the building vector data, wherein:
volume ratio FAR is total building area/block area
Building density BD is total area of building base per block area
The average building floor MF is equal to the total building area of the block/the total area of the building base of the block
Fig. 3, 4, and 5 show the calculated basic unit volume ratios FAR, the calculated building densities BD, and the calculated average number of building floors MF.
(2) Calculating the greenbelt rate by using the land data; the land use data is from ESAWorldcover2020 which has recorded land use to an accuracy of 10m globally 2020 by remote sensing technology. The boundary of the greenfield vector in the research area is extracted by introducing the boundary into the arcgis, and then the greenfield rate is calculated as follows: greenfield ratio GR is the street area greenfield area/street area.
The calculated greenfield ratios GR of the respective basic units are shown in fig. 6.
(3) Calculating a green sight rate and sky openness by using street view image data; in the embodiment, the street view image data is from a Baidu API (application programming interface), and one image is acquired at the road interval of 100m of each basic unit by writing python codes, so that 1870 street view images are acquired in total. Identifying by utilizing open-source streetscape image semantic segmentation software CUG.HPSCIL, and calculating the green vision rate and the sky openness, wherein:
green visibility GER is the number of pixels associated with a plant in an image/total number of pixels in an image
Sky openness SR (number of sky pixels in image/total number of pixels in image)
The calculated green visibility rates GER and the sky openness SR of the respective basic units are shown in fig. 7 and 8.
(4) Calculating function mixing degree and function density by using POI interest point data; the POI interest point data of the embodiment is a high-level map, and is obtained by writing python codes, wherein 14 types are recorded, and 46395 POI points are recorded, wherein the 14 types comprise: catering, shopping, transportation facilities, business housing, living services, healthcare services, sports leisure services, lodging services, scenic spots, public facilities, financial insurance services, government agencies and social groups, corporate enterprises and scientific and cultural services. Calculating the function mixedness MU and the function density FD by using the obtained POI data, wherein:
degree of functional mixing
Where n is the number of POI types in the basic cell, PiIs the probability of the type of i POI in the plot, each elementary cell Sigma Pi＝1
Function Density FD-Total number of POIs/area of neighborhood
Fig. 9 and 10 show the calculated function mixedness MU and function density FD of each basic unit.
(5) The OSM traffic network data is used to calculate the street density, where RD is the total length of the street/area, and the calculated density RD is shown in fig. 11.
(6) Drawing vector points of the city center and the subway station in arcgis according to the public activity center system planning of Nanjing city and the rail transit planning, calculating the shortest straight-line distance between the vector points and the block centroid, and obtaining the distance DC between each basic unit and the city center and the distance DM between each basic unit and the subway station as shown in figures 12 and 13.
7. General vitality of city ViAs dependent variables, 11 constructed environment indexes are used as independent variables, and a multiple linear regression model A is constructed in a machine learning library scimit-lean of python, namely 70% of 151 basic units are used as a training set, 30% of 151 basic units are used as a test set, and random _ state is 20.
8. Results are shown in Table 1, training set R of multiple linear regression model A20.659, test set R20.457 Table 1 GBDT multiple linear regression model results Table
9. General vitality of city ViAs dependent variables, 11 constructed environment indexes are used as independent variables, a GBDT regression model B is constructed in a machine learning library scinit-leann of python, namely 70% of 151 basic units are used as a training set,30% as test set, learning _ rate 0.01, n _ estimators 500, max _ depth 18, min _ samples _ split 7, min _ samples _ leaf 2, subsample 0.4, loss "squared _ error", and random _ state 20.
10. The results are shown in Table 2, where the training set RMSE for GBDT regression model B is 0.029, R20.947, test set RMSE 0.057, R2Is 0.701
TABLE 2 GBDT regression model results Table
11. As can be seen, the RSME of GBDT regression model B on the test set is much lower than that of multiple linear regression model A, while R2The method is much higher than the multiple linear regression model A, and the method obtains higher accuracy in predicting the urban block vitality
The prediction method of the present invention may be embodied in a computer-readable storage medium in the form of instructions that, when executed, implement the control method described in the above example. More specifically, the instructions may be in a computer readable language. The computer may be a general purpose computing device or a special purpose computing device. In a specific implementation, the computer may be a desktop computer, a laptop computer, a network server, a palmtop computer, a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The storage medium may be any available medium that can be accessed by a computer or a data storage device including one or more integrated servers, data centers, and the like. For example, the storage medium may be, but is not limited to, a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital versatile disk), or a semiconductor medium (e.g., solid state disk). The terminals of the above examples may be machines, devices, agents; an agent may refer to an instrument, controller, control system, etc. having automatic or semi-automatic control capabilities. The automatic control or semi-automatic control capability comes from a built-in control module, a storage medium carrying a control program, instructions or algorithms and the like.
The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention as defined in the appended claims.
Claims (10)
1. A city block vitality prediction method is characterized by comprising the following steps:
acquiring a vector boundary of a block in a region to be detected from road network data, and taking the region in the vector boundary as a basic unit;
acquiring built environment data of a block in a region to be detected from a basic unit as an independent variable;
acquiring crowd activity intensity data of a block in a region to be detected, and quantifying the crowd activity intensity data to acquire a city activity value;
matching the built environment data of the basic unit and the blocks in the area to be detected with the crowd activity intensity data to construct a training database; wherein the input sample is an independent variable, and the output sample is an urban vitality value;
training the training database by adopting a GBDT regression algorithm to obtain a city block vitality prediction model;
and predicting the block vitality of the area to be detected based on the city block vitality prediction model.
2. The city block vitality prediction method of claim 1, wherein the crowd activity intensity data comprises one or more of thermodynamic diagram data, nighttime lighting data, or social software check-in data.
3. The city block vitality prediction method of claim 2, wherein the step of quantifying the crowd activity intensity data to obtain the city vitality value comprises the steps of:
acquiring remote sensing image map data and thermodynamic diagram data of a region to be detected; carrying out geographic registration on the acquired thermodynamic diagram and the remote sensing image map data by taking the remote sensing image map data as a reference; converting the pixel value of the registered thermodynamic diagram fourth wave band data into an integer, and then converting the thermodynamic diagram fourth wave band data into vector data; finally, summarizing and counting the fourth band value of each block in the region to be measured to obtain the block heat force value HiStreet heating power value HiNamely the city vitality value.
4. The city block vitality prediction method of claim 2, wherein the step of quantifying the crowd activity intensity data to obtain the city vitality value comprises the steps of:
acquiring a night light image map of the area to be detected, and performing grid cutting on the night light image map according to the vector boundary of the area to be detected; converting the pixel value of the first wave band of the clipped night light image map into an integer, and then converting the first wave band data of the night light image map into vector data; finally summarizing and counting the first waveband numerical value of each block night light image map in the area to be detected to obtain a block brightness value NiLuminance value N of blockiNamely the city vitality value.
5. The city block vitality prediction method of claim 2, wherein the step of quantifying the crowd activity intensity data to obtain the city vitality value comprises the steps of:
acquiring social software sign-in data of an area to be detected, and screening repeated data with the same ID, the same time and the same position from the social software sign-in data; counting the check-in times F of each block in the region to be detected from the screened social software check-in dataiNumber of check-in FiNamely the city vitality value.
6. The method of claim 1, wherein the step of training the prediction model of city block activity by using GBDT regression algorithm on the training database comprises the following steps:
input training data set T { (x)1,y1,),(x2,y1,),…,(xN,yN,)},xN＝(xn,1,xn,2,…xn,k) (ii) a Where N represents the number of blocks in the area of investigation, xNThe built environment value representing the Nth block in the region to be measured in the training data set has xn,1,xn,2… et al total k independent variables, yNRepresenting the city vitality value of the Nth block in the region to be tested in the training data set;
selecting a loss function L, initializing a weak learner f0(x) Is a constant c, which should minimize the loss function L:
in the formula YiThe city vitality value of the ith block in the area to be measured;
training 1,2, …, m weak learners in sequence, and when training the m weak learners, obtaining a negative gradient value of the sample at the moment:
fitting the negative gradient value to generate a new weak learner to obtain a leaf node region R of the weak learnermjJ — 1,2 …, J — 1,2 …, J, calculated:
updating the model to obtain fm(x)：
And (3) stopping training when the Mth weak learner is trained, and finally obtaining a strong learner gradient promotion decision tree, namely a city block vitality prediction model:
7. the city block vitality prediction method of claim 1, wherein the independent variables comprise one or more of building density, volume rate, greenfield rate, function density, function mixedness, plot diversity, distance to nearest bus stop, distance to nearest subway stop, distance to city center, block area, block compactness, average building footprint.
8. The city block vitality prediction method of claim 1, wherein the training database is optimized using feature engineering operations, comprising the steps of:
preprocessing data in a training database, and performing feature conversion on preprocessed built environment data;
scoring each feature according to the correlation by using a variance selection method in a filtering type, and setting a threshold value to select the feature;
and constructing the features by adopting a cross feature method, and verifying whether the selected features can improve the accuracy of prediction.
9. A storage medium having stored thereon a plurality of programs, wherein the programs are configured to be loaded and executed by a processor to implement a city block vitality prediction method according to any one of claims 1 to 8.
10. A city block vitality prediction terminal comprising a processor adapted to execute various programs, wherein the programs are loaded and executed by the processor to implement the city block vitality prediction method of any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210351848.3A CN114662774B (en) | 2022-04-02 | 2022-04-02 | Urban block vitality prediction method, storage medium and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210351848.3A CN114662774B (en) | 2022-04-02 | 2022-04-02 | Urban block vitality prediction method, storage medium and terminal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114662774A true CN114662774A (en) | 2022-06-24 |
| CN114662774B CN114662774B (en) | 2024-07-12 |
Family
ID=82036200
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210351848.3A Active CN114662774B (en) | 2022-04-02 | 2022-04-02 | Urban block vitality prediction method, storage medium and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114662774B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115146990A (en) * | 2022-07-11 | 2022-10-04 | 南京大学 | Urban vitality quantitative evaluation method integrating multi-source geographic big data |
| CN116882831A (en) * | 2023-07-17 | 2023-10-13 | 苏州科技大学 | Urban historical cultural neighborhood public space vitality evaluation method and system |
| CN117994641A (en) * | 2024-04-07 | 2024-05-07 | 东南大学 | Automatic data acquisition method and device for urban space vitality assessment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112561401A (en) * | 2020-12-29 | 2021-03-26 | 中国地质大学（武汉） | City vitality measurement and characterization method and system based on multi-source big data |
| CN112733782A (en) * | 2021-01-20 | 2021-04-30 | 中国科学院地理科学与资源研究所 | Urban functional area identification method based on road network, storage medium and electronic equipment |
| CN112819319A (en) * | 2021-01-29 | 2021-05-18 | 华南理工大学 | Method for measuring correlation between city vitality and spatial social characteristics and application |
-
2022
- 2022-04-02 CN CN202210351848.3A patent/CN114662774B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112561401A (en) * | 2020-12-29 | 2021-03-26 | 中国地质大学（武汉） | City vitality measurement and characterization method and system based on multi-source big data |
| CN112733782A (en) * | 2021-01-20 | 2021-04-30 | 中国科学院地理科学与资源研究所 | Urban functional area identification method based on road network, storage medium and electronic equipment |
| CN112819319A (en) * | 2021-01-29 | 2021-05-18 | 华南理工大学 | Method for measuring correlation between city vitality and spatial social characteristics and application |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115146990A (en) * | 2022-07-11 | 2022-10-04 | 南京大学 | Urban vitality quantitative evaluation method integrating multi-source geographic big data |
| CN115146990B (en) * | 2022-07-11 | 2024-02-27 | 南京大学 | Urban activity quantitative evaluation method integrating multisource geographic big data |
| CN116882831A (en) * | 2023-07-17 | 2023-10-13 | 苏州科技大学 | Urban historical cultural neighborhood public space vitality evaluation method and system |
| CN117994641A (en) * | 2024-04-07 | 2024-05-07 | 东南大学 | Automatic data acquisition method and device for urban space vitality assessment |
| CN117994641B (en) * | 2024-04-07 | 2024-06-11 | 东南大学 | Automatic data acquisition method and device for urban space vitality assessment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114662774B (en) | 2024-07-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113011925B (en) | Commercial space quality evaluation method, system, equipment and medium based on big data | |
| CN112949413B (en) | City landscape element classification and locality measurement method based on street view picture | |
| CN114662774B (en) | Urban block vitality prediction method, storage medium and terminal | |
| CN112966899B (en) | Urban public service facility construction decision method influencing population density | |
| CN110458333A (en) | A kind of population spatial distribution prediction technique and system based on POIs data | |
| CN117079148B (en) | Urban functional area identification method, device, equipment and medium | |
| CN112949997A (en) | System and method for community portrayal in urban planning design | |
| CN118014297A (en) | Intelligent evaluation method and system for supply and demand responsiveness of outdoor fitness facility | |
| CN116415756B (en) | Urban virtual scene experience management system based on VR technology | |
| CN116451963B (en) | Multi-factor pension service facility optimal configuration method based on ensemble learning | |
| CN116307857A (en) | Built-up area environment evaluation method based on remote sensing image scene understanding | |
| CN114139827B (en) | Intelligent perception and optimization method for urban functional area function performance | |
| CN112836181B (en) | River light pollution index extraction method based on noctilucent remote sensing image | |
| Zheng et al. | Discovering urban functional regions with call detail records and points of interest: A case study of Guangzhou city | |
| CN115146990A (en) | Urban vitality quantitative evaluation method integrating multi-source geographic big data | |
| CN111639805B (en) | GIS-based refuse landfill site selection method | |
| CN114881309A (en) | Method for measuring characteristic correlation between urban vitality and carbon emptying | |
| Xiao et al. | Trip generation prediction based on the convolutional neural network-multidimensional long-short term memory neural network model at grid cell scale | |
| CN114861975A (en) | Urban tourism traffic demand joint prediction method based on attraction strength | |
| Li et al. | Identifying urban form typologies in seoul with mixture model based clustering | |
| CN113408867A (en) | City burglary crime risk assessment method based on mobile phone user and POI data | |
| Dong et al. | The spectral dimension of human mobility | |
| Bhandari | Urban change monitoring using GIS and remote sensing tools in Kathmandu valley (Nepal) | |
| Yu et al. | Revitalising the Silk Road: Evidence from Railway Infrastructure Investments in Northwest China | |
| CN113343781B (en) | City functional area identification method using remote sensing data and taxi track data |
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 |