top of page
NEWS


Robust Evaluation of Neural Networks for Contrail Detection
Accurate detection of aircraft contrails from satellite imagery is a valuable capability for studying one of the most impactful yet uncertain climate effects of aviation. In a recent study published in IEEE Transactions on Geoscience and Remote Sensing, Irene Ortiz Abuja and co-authors critically assess the performance of state-of-the-art deep learning models trained on the OpenContrails dataset, with the aim of explaining the segmentation performance plateau widely reported
Jan 162 min read
Â


Enhancing GOES-16 Contrail Segmentation through Ensemble Neural Network Modeling and Optical Flow Corrections
UC3M researcher Irene Ortiz spearheads pioneering work on the Contrail Detection and Segmentation Ensemble Model (CDSEM), an advanced AI...
Feb 8, 20252 min read
Â
bottom of page