• A
  • A
  • A
  • ABC
  • ABC
  • ABC
  • А
  • А
  • А
  • А
  • А
Regular version of the site

Student
Title
Supervisor
Faculty
Educational Programme
Final Grade
Year of Graduation
Artur Lukin
Software System for Local Climate Zone Classification of Urban Areas
Software Engineering
(Bachelor’s programme)
2019
This paper presents a distributed system for the classification of local climatic zones of urban areas. Along with the increasing importance of the role of cities in society and the increase in the area of urban areas, the negative impact of the growth of urban areas on the environment and the urban population has also increased. For this reason, the importance of competent planning of urbanized areas has increased, for which the concept of local climatic zones is used. Local climatic zones represent a unified classification of urban areas, which is based on the morphology of ground cover with different climatic properties and steppes of influence on the phenomenon of urban heat island. However, today the process of building maps of classes of local climatic zones requires extensive manual human labor. The system uses data of remote sensing of the Earth, in particular multispectral scenes of passive remote sensing satellites of the Earth's surface of NASA Landsat-8 and ESA Sentinel-2 missions, as well as data of active sensing of the ESA Sentinel-1 mission satellites. The main feature of the software system is the use of Apache Spark in conjunction with the GeoTreilis and RasterFrames libraries for processing large amounts of remote sensing data, which allows to accelerate the classification process of local climatic zones. For classification, the Random Forest machine learning algorithm, which is part of Spark MLlib, is used. The software system is a multi-module Maven project that includes Spring Boot applications. The scheduled software system automatically searches for and downloads NASA and ESA mission data updates, prepares multi-sensor data, including the tasks of atmospheric correction and deletion of scene areas hidden by clouds, and also calculates additional graphical indicators in order to increase the classification accuracy. The results of the classification of local climatic zones are collected in the PostgreSQL with the extension PostGIS for working with geospatial data, which in turn are available to users through a web service. The software system was developed in Java and Scala, using the Spring framework, the native GDAL library for local data processing, as well as Leaflet maps and the Apache Abdera, GeoTools, Sen2Cor libraries.

The paper contains 74 pages, 3 chapters, 5 tables, 65 figures and 3 schemes, 44 sources and 4 applications.

Keywords: geographic information system, local climate zones, earth remote sensing, machine learning, big data, Apache Spark

Student Theses at HSE must be completed in accordance with the University Rules and regulations specified by each educational programme.

Summaries of all theses must be published and made freely available on the HSE website.

The full text of a thesis can be published in open access on the HSE website only if the authoring student (copyright holder) agrees, or, if the thesis was written by a team of students, if all the co-authors (copyright holders) agree. After a thesis is published on the HSE website, it obtains the status of an online publication.

Student theses are objects of copyright and their use is subject to limitations in accordance with the Russian Federation’s law on intellectual property.

In the event that a thesis is quoted or otherwise used, reference to the author’s name and the source of quotation is required.

Search all student theses