Multisource big spatial data analysis

The field of earth observation (EO), or remote sensing, is now facing significant challenges in the processing of image data for end user purposes because of the rapidly escalating numbers of missions and sensors, and because of the range of different types of sensor being orbited. Added to this complexity is the massively increasing data archives, which themselves are regularly used for time series studies in many fields of application. An inescapable recent development in EO is the rise in the importance of citizen-generated data, often delivered over social networks. Such social media data is particularly important in emergency management and cannot be ignored in any future-focussed study looking at image interpretation in the EO context. We are now in an era of big EO data in which existing analytical algorithms will suffer performance limitations, for example, they do not scale efficiently to these new big data sets. The projects can be in one of the areas: multi-resolution analysis, artificial intelligence and spatial processing, data fusion to increase the capacity of target detection, or feature learning and feature extraction to provide good separabilty between the target of interest and the background.