| | |
Summary: NorMaL: Non-compact Markovian Likelihood for Change Detection
Osman G. Sezer
, Joseph L. Mundy
, Yucel Altunbasak
, David B. Cooper
Center for Signal and Image Processing, Georgia Inst. of Technology, Atlanta, GA
Laboratory for Engineering Man/Machine Systems, Brown University, Providence, RI
{osman, yucel}@ece.gatech.edu, {mundy, cooper}@lems.brown.edu
Abstract
This paper presents a new normalcy model of a scene
for change detection using images taken from multiple
views and varying illumination conditions. Each co-
registered pixel site is statistically modeled by a prob-
ability distribution conditioned on a set of pixels in a
non-local neighborhood that are less likely to be affected
by a change that happens at the pixel of interest. These
"non-compact neighbors" are located using information
theoretic approaches. The associated change detection
|
