ENGAGE aims to make a transformative impact on the sciences enabling non-programming scientists to create systems that semi-automatically detect objects and events in their vast quantities of audio and visual data.
By working together across two parallel, highly interconnected streams of ICT research, we will develop the foundations of statistical methodology, algorithms and systems for Interactive Machine Learning.
As an exemplar, ENGAGE partners with world leading scientists grappling with the challenge of analysing enormous quantities of data being generated in Biodiversity Science.
A member of the Zooniverse family, Bat Detective is an online citizen science project inviting the general public to contribute to biodiversity science. It is enlisting the public’s help to analyse millions of audio recordings, aiding Eurasian bat population monitoring.
Amongst the inevitable noise, their recordings contain a variety of mechanical sounds, insect calls, and bat vocalisations; automatically identifying the calls of interest is not a trivial task.
Volunteers have already identified hundreds of thousands of sounds, enabling ENGAGE to work with the Bat Detective project towards an automated classification tool for recognising bioacoustic signals in noisy, real-world, audio data.
In collaboration with ZSL, we are classifying the substrate types and organisms present in underwater imagery captured off the coast of Greenland.