Projects

To streamline the process of monitoring emissions above Thailand and the surrounding areas, this project creates an Python pipeline to automatically download, process, and visualize Sentinel-5P data. This project was conducted at the Asian Institute of Technology’s Geoinformatics Center.

Abstract # Through growingly accessible computing power and frameworks, deep learning has emerged as a promising approach to automatically extract information from images, thereby saving time and resources. This project combines earth observation (EO) data analysis with deep learning to highlight changes in the city of Cologne’s road network between the years 1998 and 2019.

This project performs a complete geospatial and semantic analysis of social media data. The methodology used has many real-world application contexts. The code used for this analysis has been implemented as a Jupyter Notebook and can also be accessed through Google Colaboratory or GitHub.

Abstract # This project provides a Spatial Data Infrastructure (SDI) that allows for the analysis and visualization of annual domestic migration flows within Germany. We utilize statistical migration data from the GENESIS-Online platform and federal-state boundary data from Eurostat to map these movements.

The following showcases a Java program developed to integrate data from a CSV file and an OGC WMS into a KML file and open the file with Google Earth. The final KML file displays tweets as placemarks along a timescale and uses a WMS image as a ground overlay.

The use of a WebGIS is of great help for visualizing and organizing spatial and non-spatial data when planning and managing large events. It can allow visitors to easily access data and query information through an application on their phone.

For a PDF version of the paper click here 1 Introduction # In the era of big data, the amount and availability of remote sensing and earth observation (EO) data as a whole has seen a large increase (Culberg & Fuhs, 2017; Lv & Wang, 2020).

This analysis is based on a fictional scenario, where UAV photogrammetry is used to determine whether the goals of a football pitch are equal in size. To do this, images of the pitch are captured and analyzed.

Located in central Europe, the city of Salzburg is home to many different species of trees. This project aims to analyze the distribution of tree species within the city. To obtain data for this analysis, the location, species, and selected attributes of trees are manually collected in the field using Esri Field Maps.

Today the capture of GNSS data has become accessible to most people via sensors built into smartphones. Although smartphone sensors are not known to have particularly high accuracy, they provide a simple, cost-effective solution for GNSS data collection.

The objective of this project is to develop a web-based mapping application that enables users to investigate the distribution of different soil types across Austria. The application provides users with information about the coverage of each soil type within Austria.

This exercise employs a variety of Object-Based Image Analysis (OBIA) Methods to classify a generated scene (Figure 1). This includes the use of sample-based and knowledge-based (including hierarchical) approaches. All practical tasks in this exercise were conducted using eCognition 9.

This project aims to perform a multiclass classification with machine learning methods to classify the quality of different wines depending on certain other attributes (e.g., pH, acidity, sugar, alcohol). Multiple steps of exploratory data analysis, preprocessing, and classification are conducted.

The following showcases a Java program that retrieves and parses features from a Web Feature Service (WFS) using GeoTools. The chosen data source is the city of Vienna’s WFS for open data.

This lab explores potential sunshine energy by conducting a limited solar energy analysis of the town Zell am See in Austria, using ArcGIS Pro for the analysis and ArcGIS Online for the presentation of the results.

In this analysis, the two methods Line of sight and Viewshed have been successfully applied to simulate a ficticious scenario for a visibility analysis. ArcGIS Pro and ArcGIS Online were utilized to carry out these analyses and to visualize and interpret the results.

In this analysis, different terrain analysis tools and techniques are explored, including slope and curvature analysis. ArcGIS Pro and ArcGIS Online were utilized to carry out these analyses and to visualize and interpret the results.

Spatial interpolation is based on the idea that nearby locations are more similar to each other than distant locations, as stated in Tobler’s First Law of Geography: “Everything is related to everything else, but near things are more related than distant things.

The following showcases a Java program that generates and opens a Google Earth KML file with a set of placemarks, each corresponding to a geographic location specified by a pair of latitude and longitude coordinates.