Advances in Hyperspectral Image Processing Techniques

9Hyperspectral Image Classification in Compressively Sensed Band Domain

Charles J. Della-Porta¹ and Chein-I Chang^1,2

¹Remote Sensing Signal and Image Processing Laboratory, Department of Computer Science and Electrical Engineering, University of Maryland Baltimore County, Baltimore, MD, USA

²Center for Hyperspectral Imaging in Remote Sensing (CHIRS), Information and Technology College, Dalian Maritime University, Dalian, China

9.1 Introduction

Hyperspectral sensing technology has found success in a variety of applications ranging from agricultural land cover/use mapping, food inspection, environmental monitoring to medical imaging, law enforcement, and military reconnaissance/surveillance. Although it has continued to improve over the years, its applications are still limited due to size, weight, and power (SWaP) constraints. One of the challenging requirements is a need of sampling a large number of very fine spectral bands which require very fast and expensive analog‐to‐digital converters, high‐capacity onboard storage, and optimized computational hardware and software to allow for real‐time processing. Such requirements limit the utility of many applications and preclude the use of hyperspectral technology in many hyperspectral data exploitation including future space hyperspectral data processing.

Compressive sensing (CS) [1] has recently developed as a promising approach in hyperspectral data acquisition and analysis. CS approaches allow for data to be acquired directly ...