next up previous
Next: Estimation of Scene Up: Probabilistic Image Fusion Previous: Introduction

Image Formation Model

Our image formation model approximates the production of the sensor images from an underlying true scene. It describes the graylevel intensities in the sensor images as locally affine transformations of the graylevel intensities arising from the scene. The model is given by

equation8

where,

tex2html_wrap_inline446 indexes the sensors,
a is the sensor image,
s is the true scene,
tex2html_wrap_inline452 is the sensor bias (captures effects of dysfunctional sensor elements,
tex2html_wrap_inline454 is the sensor gain (which includes the effects of local polarity reversals), and
tex2html_wrap_inline456 is the sensor noise

The model is defined at each location (i.e., hyperpixel) of a Laplacian pyramid. The Laplacian pyramid is a multiresolution and multiscale representation of an image. Defining the model in the pyramid domain allows one to capture the relationships between graylevel features in the sensor images and the scene existing at multiple scales.



Ravi Sharma
Wed May 12 00:15:26 PDT 1999