GRSS Lecture Series

GRSS Lecture Series
Mon, 24 July, 201704:04 PM, EDT (20:04, UTC)

This collection of lectures is presented by the IEEE Geoscience and Remote Sensing Society (GRSS). Society members enjoy free access to these videos. To learn more about GRSS membership, please visit http://www.grss-ieee.org/about/membership/

Other resources from GRSS, such as slides and tutorials, are available at their resource center.

Statistical Information Theory and Geometry of SAR Image Analysis

Statistical Information Theory and Geometry of SAR Image Analysis01:04:09
0 views

Statistical Information Theory and Geometry of SAR Image Analysis

Statistics has a prominent role in SAR - Synthetic Aperture Radar image processing and analysis. More often than not, these data cannot be described by the usual additive Gaussian noise model. Rather than that, a multiplicative signal-dependent model adequately models the observations. After summarizing the main distributions for both the univariate (intensity and amplitude) and multivariate (fully polarimetric) image formats, we present eight seemingly different problems, and how they can be formulated and solved in an unified manner from a statistical viewpoint using Information Theory and Information Geometry. A presentation by Dr. Alejandro C. Frery, from the IEEE Geoscience & Remote Sensing Society (GRSS) lecture series, originally broadcast live on IEEE.tv.

Advanced Neural Adaptive Processing in Interferometric and Polarimetric Radar Imaging - Akira Hirose

Advanced Neural Adaptive Processing in Interferometric and Polarimetric Radar Imaging - Akira Hirose01:12:08
23 views

Advanced Neural Adaptive Processing in Interferometric and Polarimetric Radar Imaging - Akira Hirose

This talk presents and discusses advanced neural networks by focusing on complex-valued neural networks (CVNNs) and their applications in the remote sensing and imaging fields. CVNNs are suitable for adaptive processing of complex-amplitude information. Since active remote sensing deals with coherent electromagnetic wave, we can expect CVNNs to work more effectively than conventional neural networks or other adaptive methods in real-number space. Quaternion (or Hypercomplex-valued) neural networks are also discussed in relation to polarization information processing. The beginning half of the Talk is devoted to presentation of the basic idea, overall framework, and fundamental treatment in the CVNNs. We discuss the processing dynamics of Hebbian rule, back-propagation learning, and self-organizing map in the complex domain. The latter half shows some examples of CVNN processing in the geoscience and remote sensing society (GRSS) fields. Namely, we present distortion reduction in phase unwrapping to generate digital elevation model (DEM) from the data obtained by interferometric synthetic aperture radar (InSAR). In polarization SAR (PolSAR), we apply quaternion networks for adaptive classification. Another example is ground penetrating radar (GPR) to visualize underground objects to distinguish specific targets in high-clutter situation. Finally we discuss the prospect of the CVNNs in the GRSS fields.