BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Random Projections for Dimensionality Reduction of Hyperspectral D
 ata
DTSTART:20111206T160000
DTSTAMP:20260407T020750Z
UID:244c261022e2affb8b2a65c48c11cec938497d40c33d96aada817c11
CATEGORIES:Conferences - Seminars
DESCRIPTION:Dr James E. Fowler\, Mississippi State University\nPrincipal c
 omponent analysis (PCA) is often central to dimensionality reduction and c
 ompression in many applications\, yet its data-dependent nature as a trans
 form computed via expensive eigendecomposition often hinders its use in se
 verely resource-constrained settings such as satellite-borne sensors. A pr
 ocess is presented that effectively shifts the computational burden of PCA
  from the resource-constrained sensor to a presumably more capable base-st
 ation receiver. The proposed approach\, compressive-projection PCA (CPPCA)
 \, is driven by projections at the sensor onto lower-dimensional subspaces
  chosen at random\, while the CPPCA reconstruction\, given only these rand
 om projections\, recovers not only the coefficients associated with the PC
 A transform\, but also an approximation to the PCA transform basis itself.
  This latter approximation is driven by a novel eigenvector reconstruction
  based on a convex-set optimization driven by Ritz vectors within the proj
 ected subspaces. The performance of CPPCA reconstruction is considered in 
 the specific application in which random projections effectuate spectral d
 imensionality reduction of hyperspectral data. In particular\, the effect 
 of such random projections on the preservation of anomalous data is invest
 igated. The popular RX anomaly detector is derived for the case in which g
 lobal anomalies are to be identified directly in the random-projection dom
 ain\, and it is determined via both random simulation as well as empirical
  observation that strongly anomalous vectors are likely to be identifiable
  by the projection-domain RX detector even in low-dimensional projections.
  Finally\, a CPPCA-based reconstruction procedure for hyperspectral imager
 y is developed wherein projection-domain anomaly detection is employed to 
 partition the dataset\, permitting anomaly and normal pixel classes to be 
 reconstructed separately in order to improve the representation of the ano
 maly pixels. 
LOCATION:ELD 220
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR