University of Nottingham Ningbo China
  • Intranet

Habitat Knowledge Discovery from Crowd-Sourced and Remotely sensed Geospatial Big Data

15 May 2013 (12:30)

About the talk

Rapid economic development for the past one hundred years has destroyed natural habitats and endangered many species. For all our technology, we rely on the delicate ecological balance of the natural world for our survival. Protecting and preserving natural habitats is therefore essential for a sustainable society. In collaboration with Ordinance Survey, the national mapping agency of United Kingdom, we are developing new technologies for automating the process of extracting meaningful habitat information from crowd-sourced and location-aware device sensed big data such as geo-referenced ground photos, aerial photos and remotely sensed images.

In this talk, I will present a new computer habitat classification method based on automatically tagging geo-referenced ground photographs.  We first present a geo-referenced habitat image database containing high-resolution ground photographs that have been manually annotated by experts based on a hierarchical habitat classification scheme widely used by Ecologists. This will be the first publicly available image database specifically designed for the development of big data analysis techniques for ecological (habitat classification) applications. We then present a novel random-forest based machine learning method for annotating an image with the habitat categories it contains. We will show how machine learning and image analysis can help processing big data emerged from crowd-sourcing, ubiquitous and location-aware computing and remote sensing to extract useful new knowledge for monitoring and protection biodiversity.


Professor Guoping Qiu is a Chair of Digital Technology and Head of School of Computer Science at the University of Nottingham Ningbo China (UNNC). In the University of Nottingham Ningbo China, he is leading the Digital Economy Research Programme at the International Doctoral Innovation Centre (IDIC), which is dedicated to researching, promoting and championing new digital technologies and solutions addressing challenges in big data, smart city, interactive new media, healthcare informatics, transport, logistics, energy and environment.

He joined the School of Computer Science at the University of Nottingham, Nottingham, UK in 2000 and for the past 20 years, he has taught in universities in the UK and Hong Kong and also consulted for multinational companies in Europe, Hong Kong and China. His research interests include image processing, pattern recognition, multimedia signal analysis, big data and digital economy. He has published widely in areas including content-based image indexing and retrieval, example-based/leaning-based image resolution enhancement and high dynamic range imaging. He also holds European and US patents.