Blue eyes: New tools for monitoring coastal environments using remotely piloted aircraft and machine learning
Remote sensing of coastal environments is a rapidly growing field of research. This is due to the use of Remotely Piloted Aircrafts (RPAs) equipped with RGB, NIR and Multi-Spectral sensors/cameras, allowing:
• Rapid deployment of such aircrafts,
• High-resolution images of entire landscapes and environmental features within them.
This opens up the possibility to employ image processing techniques to further analyse possible patterns in data. The use of RPAs and subsequent captured data from coastal environments allows for mapping and tracking features, such as: vegetation, shellfish, birds, mammals, geomorphology and even litter. This also allows time-profiling of certain environments to be performed. The use of RPAs to captures 2D data at multiple points of view which may allow for 3D information to be deduced (structure from motion). This project will focus on developing tools to speed-up the image analysis process of mapping and tracking environmental features, given recent developments to the field of Deep Learning and to explore 3D imaging process from multiple 2D images.