Delimitation of Urban Areas with Use of the Plataform Google

Engine Explorer

Sherlyê Francisco de Carvalho

, Jhonatha Fiorio Conceição Guimarães

Carla Bernadete Madureira Cruz

and Elizabeth Maria Feitosa da Rocha de Souza

Institute of Geosciences, Rio de Janeiro Federal University, Computer and Mathematical Science, Rio de Janeiro, Brazil

Department of Geography, Rio de Janeiro Federal University, Rio de Janeiro, Brazil

Keywords: Google Engine Explorer, Urban Expansion, Mapping Urban, Remote Sensing.

Abstract: Google Earth Engine is a cloud computing platform for developing and hosting web applications that allows

for the automatic sorting and mapping of terrestrial coverage. The objective this research is to evaluate the

tool's potential for the generation of thematic maps of urban areas, using a big data plataform, with data in

the cloud. The proposed methodology evaluate the CART classifier for different scales. The local scale

considered the area of Rio de Janeiro city. A simplified legend (urban and non-urban) and other with greater

detailing (different types of urban intensify), were tested. The main input was the Landsat TOA (Top of

Atmosphere) mosaic. Potential, classification time, and results were evaluated. The main products generated

were the temporal classifications, in which one can observe the expansion of urban areas and some

confusion between classes. In this case editing is necessary. The rapidity in the classification and generation

of products is one of the most important positive points of the analysis. The tool is very interactive and easy

to handle, even by users with little experience. The urban areas delimitation and identification were

promising, requiring more research on the best techniques to be adopted at each geographic scale.

1 INTRODUCTION

Geography as a science deals with the analysis of

space and the actions of transforming agents of this

space.

The traditional mapping based on the use of

desktop tools, in actual moment has been

complemented by the use of online platforms for

data storage and processing.

In this context, cloud computing and the use of

Big Data are presented as a potential model for

spatial analysis, and mapping.

The Google Earth Engine, and in special the

Earth engine Explorer, is a tool with enormous

potential to create geographic analyzes, in

attending to several applications in the geosciences,

and to the Geography.

Among the potential analyzes is the

delimitation of urban areas and their expansion,

from mappings and the monitoring of these areas.

2 RELEVANCE

Given the increasing volume of data generated daily

it is important to consider the use of spatial analysis

tools that allow integration, ease of access and speed

in data processing.

The access and sharing of the web, brings in this

way, the discussion about of the Big Data concept of

and the online systems for the generation of

mappings and geographic analyzes.

The evolution of cloud computing is one of the

greatest advances in the history of computing,

becoming a new paradigm in recent years. Among

the various existing definitions, cloud computing can

be defined as a collection of virtualized and

interconnected computers that provide computing

resources and services and are dynamically

provisioned and presented based on an agreement

between the service provider and the consumer

(R.Buyya, 1999). In a simpler way, it can be

considered as a virtualization of a data center,

where servers are virtualized seeking the best

use of their resources that are made available

Carvalho, S., Guimarães, J., Madureira Cruz, C. and Feitosa da Rocha de Souza, E.

Delimitation of Urban Areas with Use of the Plataform Google Engine Explorer.

DOI: 10.5220/0006756602550261

In Proceedings of the 4th Inter national Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2018), pages 255-261

ISBN: 978-989-758-294-3

255

through virtual machines. A cloud can be

deployed publicly, with a service provider

providing the resources and services that the

organization needs, or privately, and is managed

internally by the organization. There may also be

a hybrid structure where an organization

maintains an internal infrastructure and provides

some services publicly. A public cloud is

characterized by being available through a third

party service provider via the Internet. It is a

cost-effective way to deploy an IT solution,

especially for small and medium-sized

enterprises, and government entities that need to

provide a variety of services to the population.

The leading companies that provide cloud

computing services are Amazon AWS, Microsoft

Azure and Google Cloud. In addition to these,

there are several companies that provide only

certain services, such as storage.

3 OBJECTIVES

The objective of the present work is to evaluate the

potential of the Google Earth Engine Explorer tool

aiming the generation of thematic maps for the

delimitation of urban areas using Cloud Computing,

Big Data and Remote Sensing.

- To evaluate the mapping of urban areas from

the Google Earth Engine platform using

Explorer mode, with the CART classifier, to

the limits of the Rio de Janeiro city and scale

regional too.

- Generate a mapping of urban delimitation

and expansion considering different

occupancy intensities.

4 CONCEPTUAL REVIEW

4.1 Data and Spatial Information

The given terms and information academic are used

several times with the same meaning, but in reality

they have different meanings. Given is the set of

different observations that are collected and stored.

Information is a data that is useful for answering

questions or for solving a problem, and has a

interpretation.

Two major trends have had a profound impact on

spatial data management in recent years. The first of

these is the exponential growth of the data volume.

the secund is the change of criteria for a database to

be considered to be large, and this is the result of the

wide integration of spatial information in

productions continues from many users.

Another important aspect is the current spatial

distribution of the users of geographic information

accessing contents in different places and with

differentiated platforms.

4.2 Geographic Data Base and BIG

DATA

A geodatabase is a database capable of storing,

querying and manipulating geographic information

and spatial data of any kind.

With it you can manipulate simple geometric

shapes: dots, lines and polygons, or to use features

such as topology, 3D technology, images, to better

represent real-world phenomena.

The geodatabase can be stored on top database

management systems and allows to organize and

manage a huge volume of high-performance data in

a multi-user environment.

The challenges in a geodatabase, among others,

are: analysis, capture, treatment, search, sharing,

storage, transfer, viewing and data privacy

information.

Big Data is a term that has long been used to

name very large or complex data sets that traditional

(desktop) data processing applications still can not

handle.

In other words, the ever-increasing production of

large-volume data requires new alternatives for data

management, online data sources, and the need for

processing beyond traditional methods.

Cloud computing distributes resources in the

form of services. We can then divide it into three

types: Software as a Service (SaaS), Platform as a

Service (PaaS) and Infrastructure as a Service

(IaaS).

Software as a Service (SaaS) The SaaS model

provides software systems for specific purposes,

which are accessed via the Internet by a web

browser for example. In SaaS, you do not manage or

control the infrastructure of the lower layers, except

for specific system configurations. With this,

developers focus on innovation rather than

infrastructure, leading to the rapid development of

software systems (Souza et al., 2010). user can use

vendor applications that run on a cloud

infrastructure. But the consumer does not manage or

control the basic infrastructure, including network

clouds, operating systems, storage, servers. SaaS

GISTAM 2018 - 4th International Conference on Geographical Information Systems Theory, Applications and Management

256

runs entirely on the cloud bringing cost savings, thus

dispensing with the acquisition of software licenses.

Platform-as-a-Service (PaaS) PaaS providers

provide development environments so that

developers do not have to worry about the

infrastructure that will be used, not even with the

installations of the environments used by their

applications, expensive and complex jobs in the vast

majority of cases. In PaaS, users do not have cloud

infrastructure management, but have control over the

deployed applications and the ability to configure

the hosted application environment. consumer can

deploy in the cloud the infrastructure created or the

acquired applications created using programming

languages and the tools supported by the provider. It

does not manage or control the basic infrastructure,

but controls used applications and hosting

applications and environment settings. Provides an

infrastructure to deploy and test applications in the

cloud. It also provides an operating system,

programming languages and development environ-

ments for applications, aiding the implementation of

software, as it contains development tools and

collaboration between developers.

Infrastructure as a Service (IaaS) the term IaaS,

refers to a computational infrastructure that uses

virtualization techniques to deliver computational

resources. An infrastructure in the IaaS model aims

to make it easy and affordable to manage and deliver

computing resources, that is, it is responsible for

providing resources such as servers, network,

storage and even operating systems and applications

required to build a environment on demand. In

addition to providing in most cases, online services

for infrastructure administration, such as a web

interface. Because it is the lower layer, it is also

responsible for providing the infrastructure used by

the middle and upper layers. Amazon EC2 (Cloud

Virtual Servers) and Amazon S3 (Scalable Cloud

Storage) services are examples of IaaS. Cloud

providers usually charge for the IaaS service for the

total of resources allocated or consumed (Amazon

AWS, 2016). brings the services offered at the

infrastructure layer, in these services we can include

various computing resources such as servers,

routers, storage systems, among others. Responsible

for providing all the necessary infrastructure for the

two previously mentioned structures SaaS and PaaS.

IaaS is based on computing resource virtualization

techniques. Looking from the economy side, haven't

to buy new servers or even network equipment for

service expansion, because everything that is need is

included in the cloud.

4.3 Urban Areas

The process of urbanization in Brazil began in the

twentieth century, and the industrialization led the

population to go from the countryside to the city

(rural exodus).

Urbanization is an increase of the urban

population compared to rural. The economy in

Brazil went from agrarian-exporter to urban-

industrial, and the understanding of the urban is still

based on the idea of capitalist and industrial society.

Currently, more than 80% of the Brazilian

population lives in urban areas.

Disorganized urbanization causes a series of

social and environmental problems.

The solution to environmental and social

problems is what motivates today the study of the

growth and densification of urbanization.

4.4 Google Earth Engine and the

Explorer Platform

Google Earth Engine is a cloud computing platform

for developing and hosting web applications on

Google's infrastructure that enables, among other

applications, automatic mapping and classification

of land cover.

It was initially released as a beta release in April

2008 but only in 2017 there was a massive publicity

and incentive to use by the company (in Brazil).

Pesquisando Google Earth Engine no Google

acadêmico podemos achar 20 resultados de artigos

ou publicações no período de 2008 a 2016 e 26 no

período de 2016 a 2017.

Google Engine is a technology in the Platform as

a Service (PaaS) model, which virtualizes

applications on multiple servers, providing

hardware, connectivity, operating system, and

software services. It can be used for free to a certain

level of resource consumption with numerous tools

ready and adapted for users without programming

knowledge. Additional fees may be charged for the

consumption of resources when the user has

technical knowledge to implement more complex

analyzes.

The rapidity in the classification and generation

of products is one of the most important positive

points of the analysis. The tool is very interactive

and easy to handle, even by users with little

experience.

It uses cloud-processing technology, which

enables the use and manipulation of large volumes

of georeferenced data. Its system is designed to

enable scientific analysis and visualization of

Delimitation of Urban Areas with Use of the Plataform Google Engine Explorer

257

geospatial data sets. The data catalog has an

immense volume of data and a wide range of

popular datasets, such as the world's largest

collection of Landsat scenes, 25 years of high-

resolution images, and other Landsat images since

1972, has variety of data types, bands, projection, bit

depth, spatial resolution, temporal. It has data from

the Sentinel, Images Moderate Resolution Imaging

Spectroradiometer (MODIS), night-time imagery-

Defense Meteorological Satellite Program's

Operational Linescan System (DMSP-OLS), digital

elevation models, slope data, surface temperature,

climate, atmospheric data beyond of global daily

satellite feeds.

The user can add and store their own data and

collections. Upload your own image with Maps

Engine, in the original projection, with all the bands

and metadata. In addition to being able to save your

data, collected points, classifications and these can

be used, manipulated and opened in traditional

programs.

The platform can be used in two ways, in the

"Explorer" mode a programming interface that was

used in this work, also contains the "Code Editor"

mode used by users with programming knowledge.

The platform has already been successfully used

for various purposes. Between them:

The European Commission's Joint Research

Center (JRC) has used the Earth Engine to develop

high-resolution maps of global surface water

occurrence, change, seasonality, recurrence and

transitions.

Collect Earth, developed by the Food and

Agriculture Organization of the United Nations

(FAO), is a free, open-source, easy-to-use tool using

Google Earth and Google Earth Engine to view and

analyze land lots to deforestation and other forms of

land use change.

Global Forest Watch, an initiative of the World

Resources Institute, is a dynamic online forest

monitoring system designed to enable better

management and conservation. Global Forest Watch

uses Earth Engine to measure and visualize changes

in the world's forests.

A team led by Matt Hansen of the University of

Maryland used the Earth Engine to research more

than a decade of global warming extension, loss, and

gains. This area is 128.8 million square kilometers,

equivalent to 143 billion pixels of Landsat data in a

spatial resolution of thirty meters.

5 METHODOLOGY

The proposed methodology starts with the choice of

analysis scales and inputs. The CART classifier for

the mesoscale Rio de Janeiro was initially evaluated.

Simplified legends (urban and non-urban) and

greater detailing (different types of coverage of

urban areas and levels of urban intensity) were

tested. The main input was the Landsat TOA (Top of

atmosphere) mosaic. The potential, time of

classification, and results were evaluated.

Using the Google Earth Engine platform in

"Explorer" mode for computer laymen and selecting

images from different years in the platform database,

the classifier was chosen. In this case the CART-

Classification and Regression Trees (The decision

tree method is a supervised learning approach, that

is, it comprises the abstraction of a knowledge

model from the data presented in the form of ordered

pairs (desired input and output) [Goldschmidt e

Passos 2005]. n this method, the production of the

results presents simplicity and readability for its

interpretation, fact that, according to Oliveira

(2005), has become one of the main advantages of

its use. With regard to the CART algorithm, one of

its main characteristics is the research capacity and

relations between the data, involving the

construction and simplification phases of the

decision tree, choosing the best variable for dividing

the data into two nodes, where the procedure is

applied recursively to the data in each of the child-

nodes and so on [Hand et al 2001].) (a classification

algorithm that has one of its main characteristics the

research capacity and relations between the data,

involving the phases of construction and

simplification of the decision tree, choosing the best

variable for dividing the data into two nodes, where

the division procedure is applied recursively to the

data in each one of the nodes (classification is a

process that finds common properties between a set

of records belonging to a database and classifies

them into different classes according to a model)

through the indication of samples pixel-by-pixel of

the different types of subtitles that were used.

Five tests were performed. In the first test we

tried to define the potential of delimitation of urban

areas considering only two classes (urban and water)

and others. in the second test the number of classes

was increased. were selected for vegetation, soil and

sand. In this case, the results were more promising.

In the third test, the classification for the

delimitation of urban areas on a regional scale for

the southeastern region of Brazil was enhanced. In

this case the Nigth time ligth file was used as support.

GISTAM 2018 - 4th International Conference on Geographical Information Systems Theory, Applications and Management

258

Version 4 of the DMSP-OLS Nighttime Lights

Time Series consists of cloud-free composites made

using all the available archived DMSP-OLS smooth

resolution data for calendar years. In cases where

two satellites were collecting data - two composites

were produced. The products are 30 arc second

grids, spanning -180 to 180 degrees longitude and -

65 to 75 degrees latitude.

In the final tests (4 and 5) classifications were

made to delimit different intensity levels of urban

areas. This evaluation considered the scale of detail

(minutemen of Rio de Janeiro10 and carried out for

the years 2000 and 2010, aiming at the use in the

monitoring.

After the generated products was applied Kappa

(Cohen, 1960) Gong & Howarth (1990) used the

Kappa (K) index as an important precision measure

to be associated with the error matrix, since it

represents it entirely, that is, considering all

elements of the matrix and not only those that are on

the main diagonal, as it happens with the Global

Accuracy index. The index using an image of

Google Earth itself and 200 random points to make

the conference of the classified areas. The Kappa

index is a pointer to show reliability of the

measurement procedures used which is a key issue

in any research study. It is a measure of agreement

that measures the degree of conformity beyond what

would be expected by chance alone. And with all the

finished products, the results of the images,

classifications (fig.1 and 2), tables 1 and 2, and

evaluating the Kappa index that was calculated as

shown by the equation below can be analyzed.

k=(∑x_ij-∑x_(i. )∑x_j)/(1-∑x_i⋅X_j ) (1)

Equation 1: Kappa equation.

Table 1: Kappa value and index of points.

Kappa value

Class

< 0.20

Poor

0.21 - 0.40

Low

0.41 - 0.60

Moderate

0.61 - 0.80

Good

0.81 - 1.00

Very good

Font: Adapted from Galparsoro and Fernandez (2001).

6 RESULTS AND CONCLUSIONS

6.1 Results

The table 2 shows the results to the tests 1 -4.

Table 2: Results to the tests 1 and 2.

Tests

Image/mosaic/

area and scale

Class and

number

samples

Results

(1)

LANDSAT 8

Mosaic year

(2000) - Rio de

Janeiro city

1:100000

Urban (83),

Water (71),

Other (31)

Few class the

result

overestimate

the urban

areas

(2)

LANDSAT 8

Mosaic year

(2000) - Rio de

Janeiro city

1:100000

Urban (83),

Water (71),

Vegetation (66)

Soil (64)

Beach sand(22)

Other (10)

Delimits

urban areas,

water bodies,

vegetation

with good

results.

(3)

LANDSAT 8

Mosaic year

(2000) + Night

time Lights –

regional scale

1:1000000

Urban(83),

Water(71),

Vegetation (66),

Soil (64),

Beach sand(22)

Other (10)

Better

delimitation

of the urban

areas.

(4)

LANDSAT 8

Mosaic years

(2000, 2010) -

Rio de Janeiro

city 1:100000

Intense Urban

(20),

Average urban

(22),

Rarefied urban

(24),

Vegetation (08),

Water (18)

The result

delimits well

diffetents

urban levels.

It is possible to observe through the figures 1 and 2

of the generated classifications the following results.

Figure 1: Simplified classification (test 1).

Delimitation of Urban Areas with Use of the Plataform Google Engine Explorer

259

Figure 2: Detailed classification (test 2).

Figure 3: Nigth time ligth classification - Image index

Kappa de 0,85 (test 3).

Figure 4: Image and classification of the Rio de Janeiro

city in 2000 (test 4/5).

6.2 Conclusions

About the Google Earth Engine Explorer Platform

was observed high speed in the processing of the

samples and in the generation of the result (mapping

of the cover); allows to open different tabs for the

generation of differentiated classifications; the

number of classes and samples very little affect the

Figure 5: Image and classification of the Rio de Janeiro

city in 2010 (test 4/5).

processing speed which is very high; changing the

resolution to the classification changes the result and

must consider the scale of the mapping; and the

classifier proved efficient for the delimitation of

urban areas.

On the mapping one can conclude that its use has

great potential for Geography with excellent result

of the Kappa index; it has the possibility of

applications aiming the monitoring of the urban

expansion; and the possibility of geographical

analysis at different scales.

It is necessary to evaluate the other classifiers as

well as to compare tests in other scales and

geographic areas.

Validation must consider different methods and

validate other images made available on the Google

Earth Engine Explorer platform.

REFERENCES

Cohen, J. (1960) A coefficient of agreement for nominal

scales. Educational and Psychological Measurement

20, 37-46.

Galparsoro, L.; fernández, S.P. Medidas de concordância:

el índice Kappa. 2010.

Goldschimidt, R e passos, E. Data mining: Um guia

prático. Rio de Janeiro: Campus, 2005.

Gong, P.; P. J. Howart, 1989. A modified probabilistic

Relaxation approach for land-cover classification.

IGARSS 8921th Canadian Symposium on Remote

Sensing, Vancouver, British Columbia. PP.1621-1624.

Oliveira, R. (2006) “Uso de Data Mining para obter perfis

de clientes com maior lucratividade”, Monografia

(Especialização em Gerenciamento em Banco de

Dados), Universidade do Extremo Sul Catarinense,

Criciúma.

R. Buyya (Ed.), High Performance Cluster Computing:

Architectures and Systems, vol. 1, Prentice Hall,

Upper Saddle River, USA, 1999.

Sousa, F., Moreira, L., Machado, J. Computação em

Nuvem: Conceitos, Tecnologias, Aplicações e

GISTAM 2018 - 4th International Conference on Geographical Information Systems Theory, Applications and Management

260

Desafios. In: Antônio Costa de Oliveira;Raimundo

Santos Moura;Francisco Vieira de Souza. (Org.). III

Escola Regional de Computação Ceará, Maranhão e

Piauí (ERCEMAPI). 1 ed. Teresina: SBC, 2009, v. 1,

p. 150-175.

Delimitation of Urban Areas with Use of the Plataform Google Engine Explorer

261