MSc. Thesis/Internship: Building Detection and Tracking over urban areas

Job title: Water Resources Analysis

Work Location: Germany/Remote

Department: Remote sensing, Computer Science, Earth Sciences

Work load: 30 Hours / Week

Job Type: Master Thesis

General Description:

In the frame of prototyping a Remote sensing solution that is able to process, analyze and interpret satellite data using machine learning and AI algorithms (see images below).

It is required based on the use case to train our AI model to detect, classify or predict a specific feature or phenomena for example: Trees, buildings, streets, farms, oil fields, etc.

Infact, more than 100 countries currently lack effective Civil Registration systems according to world bank estimation. On top, Quantifying population statistics is fundamental to 67 of the 232 United Nations Sustainable Development Goals. Further, it have many implications for disaster preparedness, environment, infrastructure development, and epidemic prevention. That’s why we are motivated to develop an AI solution that can detect and track building construction over time. The training data needed for this purpose is already available on Nabtaplaya AWS S3 storage. Its required to Build, train, and deploy the machine learning (ML) model for the use case with fully managed infrastructure, tools, and workflows on AWS SageMaker. Two types of training data are available. First one, is a time series of Planet imagery that includes 24 images (one per month) covering ~60 unique geographies. The dataset will comprise over 20,000 square kilometers of imagery labeled of building footprints. Second one, Over 120 sq km of both high resolution synthetic aperture radar (SAR) data and electro optical (EO) imagery with ~48,000 building footprint labels of Rotterdam, The Netherlands.

Essential Duties and Responsibilities:

• Build, Train and deploy AI models to detect and track buildings from satellite time series data

• Implement the prototype using AWS Infrastructure (S3, SageMaker, EC2)

Education and/or Work Experience Requirements:

• Bachelor in computer science, remote sensing or related disciplines

• Very good knowledge of Python

• Very good knowledge of Machine Learning and AI

• Very good knowledge of Image Processing

• Very good knowledge of Linux

• Good knowledge of Cloud Technology

• Very good verbal and written communication skills in English

• Ability to work independently and to carry out assignments to completion within parameters of instructions given, prescribed routines, and standard accepted practices