Prasad S., Bruce L.M., Chanussot J. (eds.) Optical Remote Sensing. Advances in Signal Processing and Exploitation Techniques

Springer, 2011. — 351 p. — Augmented Vision and Reality, том 3.The evolution of optical remote sensing over the past few decades has enabled the availability of rich spatial, spectral and temporal information to remote sensing analysts. Although this has opened the doors to immense possibilities for analysis of optical remotely sensed imagery, it has also necessitated advancements in signal processing and exploitation algorithms to keep up with advances in the quality and quantity of available data. As an example, the transition from multispectral to hyperspectral imagery requires conventional statistical pattern classification algorithms to be modified to effectively extract useful information from the high dimensional hyperspectral feature space. Although hyperspectral imagery is expected to provide a much detailed spectral response per pixel, conventional algorithms developed and perfected for multispectral data would often be suboptimal for hyperspectral data. At best, they would require a significant increase in the ground-truth (training) data employed for analysis—something that is often hard to come by, and is often far too costly. As a result, signal processing and pattern recognition algorithms for analysis of such data are also evolving to cope with such issues and result in practical applications.
The last decade has seen significant advances in algorithms that represent, visualize and analyze optical remotely sensed data. These advances include new algorithms to effectively compress high dimensional imagery data for efficient storage and transmission; new techniques to effectively visualize remotely sensed data; new analysis and classification techniques to analyze and classify remotely sensed imagery; and techniques to fuse remotely sensed imagery acquired simultaneously from different sensing modalities. This book brings together leading experts in these fields with the goal of bringing the cutting edge in signal processing and exploitation research closer to users and developers of remote sensing technology. This book is not intended to be a textbook for introductory remote sensing analysis. There are existing textbooks that provide a tutorial introduction to signal and image processing methods for remote sensing. This book is intended to be a valuable reference to graduate students and researchers in the academia and the industry who are interested in keeping abreast with the current state-of-the-art in signal and image processing techniques for optical remote sensing.Hyperspectral Data Compression Tradeoff
Reconstructions from Compressive Random Projections of Hyperspectral Imagery
Integrated Sensing and Processing for Hyperspectral Imagery
Color Science and Engineering for the Display of Remote Sensing Images
An Evaluation of Visualization Techniques for Remotely Sensed Hyperspectral Imagery
A Divide-and-Conquer Paradigm for Hyperspectral Classification and Target Recognition
The Evolution of the Morphological Profile: from Panchromatic to Hyperspectral Images
Decision Fusion of Multiple Classifiers for Vegetation Mapping and Monitoring Applications by Means of Hyperspectral Data
A Review of Kernel Methods in Remote Sensing Data Analysis
Exploring Nonlinear Manifold Learning for Classification of Hyperspectral Data
Recent Developments in Endmember Extraction and Spectral Unmixing
Change Detection in VHR Multispectral Images: Estimation and Reduction of Registration Noise Effects
Effects of the Spatial Enhancement of Hyperspectral Images on the Distribution of Spectral Classes
Fusion of Optical and SAR Data for Seismic Vulnerability Mapping of Buildings