Title of Document: PHYSICS-BASED DETECTION OF SUBPIXEL TARGETS IN HYPERSPECTRAL IMAGERY Joshua Bret Broadwater Doctor of Philosophy, 2007 Directed By: Professor Ramalingam Chellappa Department of Electrical and Computer Engineering Hyperspectral imagery provides the ability to detect targets that are smaller than the size of a pixel. They provide this ability by measuring the reflection and abs...

We introduce a novel class of signatures---spectral edges and end points---in 21-cm measurements resulting from interactions between the standard dark sectors. Within context kinetically mixed photon, we demonstrate how resonant photon-to-photon conversions can imprint distinctive spectral features in observed brightness temperature, with implications for current, upcoming, proposed experiments...

We have assembled a wide spectral range, allelectronic spectro-polarimetric imager utilizing an acousto-optic tunable filter (AOTF), a liquid crystal variable retardation plate, and real-time, softwareadaptable signature masks. This combination of rapidly adjustable parameters allows operation at 30/sec. frame rate, and near real time adaptability to changing target signatures. The spectral cap...

Person re-identification in a multi-camera environment is an important part of modern surveillance systems. Person reidentification from color images has been the focus of much active research, due to the numerous challenges posed with such analysis tasks, such as variations in illumination, pose and viewpoints. In this paper, we suggest that hyperspectral imagery has the potential to provide u...

In this work, a novel methodology based on hyperspectral imagery with enhanced Darkfield microscopy for probing and characterizing changes in blood cell components was tested. Two main categories of blood cells were analyzed, red and white blood cells. Unique spectral signatures of ordinary and most common deformed morphologies of red blood cells were identified. Moreover, examination of white ...

The effectiveness of many hyperspectral feature extraction algorithms involving classification (and linear spectral unmixing) are dependent on the use of spectral signature libraries. If two or more signatures are roughly similar to each other, these methods which use algorithms such as singular value decomposition (SVD) or least squares to identify the object will not work well. This especiall...

The insights gained from present land cover classification activities suggest integration of multi-angle data into classification attempts for future progress. Land cover types that exhibit distinct signatures in the space of remote sensing data facilitate unambiguous identification of cover types. In this first part, we develop a theme for consistency between cover type definitions, uniqueness...

PURPOSE: This technical note demribea the spectral and spatial characteristics of hyperspectral data and the potential application of these data for wetlands studi~ and monitoring applications. The advantages and disadvantages of these data for wetland evaluations are discussed. Spectral signatures extracted tim data squired by NASA’s collected Airborne Visible/Infked Imagery Spectrometer (AVIR...

The insights gained from present land cover classification activities suggest integration of multiangle data into classification attempts for future progress. Land cover types that exhibit distinct signatures in the space of remote sensing data facilitate unambiguous identification of cover types. In this first part, we develop a theme for consistency between cover type definitions, uniqueness ...

A data-processing technique is proposed for use with conventional frequency-chirped absorption spectroscopy to ensure accurate mapping of spectral features into time-domain signatures with arbitrarily fast readout chirp rates. This technique recovers the spectrum from a signal that is distorted owing to the fast chirp rate and therefore facilitates fast measurement of the spectral features over...