Design and implement a (Deep) Neural Network to identify and output rooftop coordinates in a specific region of interest (Ref 17/01)

Company: Bird.i

Location: 20-23 Woodside Place, Glasgow

Company Description: Bird.i is a Big Data Analytics start-up exploiting satellite, airborne and UAV imagery. Its objective is to create the market place for observation imagery, serving any type of businesses and individuals. Bird.i’s ambition is to bring observation imagery to the reach of the mass market via an API to be consumed in mobile and web applications.

Project Description: Surveying buildings to establish rooftop surface areas for solar panel installation is a time-consuming and labour-intensive effort. The aim of this project is to build a big data machine-learning pipeline that will aid in automated estimation of rooftop surface areas in a high-resolution satellite image based on raster map tiles. The solution will be trained and used on satellite images from a limited geographic area. The scope will be defined at the beginning of the project.

Student Specification:

  • Programming skills in Python
  • Knowledge of Machine Learning and Neural Networks for Image classification
  • Familiar with Deep Learning APIs like Tensorflow, Caffe and Theano
  • Familiar working in an IDE (preferably Eclipse for Python), version control and Agile methodology
  • Ability to work on their own and as a part of a team

Target courses: Big Data, Data Analytics, Remote Sensing, Earth Observation

The Nitty Gritty: Work period to be agreed with successful candidate but nominally to start 5 June 2017 for a duration of three months. £1,100 pcm.

Closing Date for Applications: This role is now closed.

Cookies on Catapult explained

To comply with EU directives we now provide detailed information about the cookies we use. To find out more about cookies on this site, what they do and how to remove them, see our information about cookies. Click OK to continue using this site.