development of web-based platform for measuring
TRANSCRIPT
1. IntroductionIn connection with the exponential increase of big data, the Internet and distributed
computing provide the technical framework for processing geospatial data. Web-based GISprovides a gateway to explore and access geographic WEB services, which help to visualize andanalyze the geographic data for GIS users. In particular, to measure and compare the urbangrowth process of big cities in a multidimensional way, the Web-GIS can inspire and assist inunderstanding the urbanization process in a more comprehensive manner.
2. Data and MethodsBoth geospatial data (three kinds of optical image data, cloud-based vector data) and
statistical data are utilized to extract the indicators. The optical imaged data utilized are listed:• Land use / cover (LULC) map: Landsat TM/ ETM+/ OLI satellite imagery.• Energy consumption intensity map: DMSP-OLS nighttime imagery.• Population data: ORNL's LandScan global population distribution data.
Development of web-based platform for measuring urbanizationHao GONG and Yuji MURAYAMA
Division of Spatial Information Science,Graduate School of Life and Environmental Sciences, University of Tsukuba
Contact email: <[email protected]> Web: http://giswin.geo.tsukuba.ac.jp/mega-cities/
D-08CSIS 2016
Fig.1: Overview of the Web-GIS platform. (http://giswin.geo.tsukuba.ac.jp/mega-cities/)
3. ResultsIn this project, an open access web-based GIS platform was established (Fig.1). 35 major cities
in Asian and African regions (Fig.3) were selected to examine the urban sprawl of mega-citiesduring the 15 years (2000-2014). In order to quantify and measure the urbanization process, allof these cities were separated into three groups (African cities, Asian cities and Chinese cities),and the size of the target area is a 100 × 100 km extent for each city. All of the processeddatasets could be downloaded from the website.
Preprocessing(Band Layer
stacking, mosaicking, extent
decision)
Optical image acquisition(Landsat,
DMSP/OLS, LandScan)
Digital image analysis with machine learning(User spatial analysis, supervised
classification, training sites, raster calculation)
Accuracy assessment
Processed results(Local storage)
Database transformation
(KML files, fusion table formats)
Cloud storage data acquisition
(Administration, road net layer, transportation layer, real-time
traffic layer)
Preprocessing(Data organization,
data grafting)
Cloud-based spatial data customization(Data reconstruction and merging,
spatial analysis )
Data embed(Web-based GIS
platform)
Data mining by GIS professional
Big data collecting
Processed results(Cloud storage)
Phase 1: (Backend development)Workflow of distributed GIS data processing
Phase 2: (Front-end development)Workflow of web-based GIS platform development
Loca
l-b
ased
Clo
ud
-b
ased
Preprocessing(Skill learning, API keys registration )
Web-GIS framework
construction
Distributed GIS data embed
(Cloud and local processed data)
Web-GISsub functions development
(Spatial analysis, data visualization,
table views, download services)
System integration(Construct
streaming data transmission
system)
Web-GIS platform release
Users(Casual users,
developers, spatial planners, open data
community)
Fig.2: Workflow of the system development.
Populationdistribution maps
Energy consumption intensity maps
Land use/cover maps
Real-time traffic maps, statistic info, search function, thematic road network
4. AcknowledgementThe assistance provided by Dr. Ronald C.
ESTOQUE, Mr. Shyamantha Subasinghe, Mr.Hao HOU, Mr. Matamyo SIMWANDA, andMr. Xinmin ZHANG of the Division of SpatialInformation Science, University of Tsukubaduring the land-use/cover classification andaccuracy assessment.
Fig.3: Summary and analysis page for all cities.
Mu
ltid
imen
sio
nal
mea
sure
men
t m
od
ule
Cit
y co
mp
aris
on
mo
du
leA
nal
ysis
mo
du
le
Dataset download module