Applications of Deep Learning to Radar Polarimetry: A Physics First Approach to Machine Learning in Radar Earth Observation Applications

De, Shaunak

ISBN 10: 6139815991 / ISBN 13: 9786139815999

Edité par LAP LAMBERT Academic Publishing, 2018

A propos de cet article

212 pages. 8.66x5.91x0.48 inches. In Stock. N° de réf. du vendeur zk6139815991

Détails bibliographiques

Titre : Applications of Deep Learning to Radar ...

Éditeur : LAP LAMBERT Academic Publishing

Date d'édition : 2018

Reliure : Paperback

Etat : Brand New

A propos de ce titre

Synopsis :

Radar remote sensing has made significant technological and scientific advances in the past few years. Sensors and constellations are able to acquire high resolution, polarimetric, wide swath data with high temporal repetivity. This has lead to an exponential increase in the volume of data available. With more temporally dense constellations planned in the near future, it is imperative that automated techniques based on machine learning algorithms be developed that are able to take advantage of all the acquired data and convert latent information to actionable knowledge. However, the use of indiscriminate machine learning techniques can be problematic since there is no guarantee that the learned model makes sense from a physical standpoint. Advanced neural network techniques, collectively called 'deep leaning' algorithms have demonstrated the ability to self-learn features from a data-volume, greatly reducing the need for time-consuming feature tuning. In this book, novel deep learning algorithms and architectures are detailed for various earth observation applications using fully polarimetric SAR data based, and constrained by the principles of scattering physics.