@ARTICLE\{IMM1998-0219,
    author       = "R. Larsen and K. Conradsen and B. k. Ersb{\o}ll",
    title        = "Estimation of Dense Image Flow Fields in Fluids",
    year         = "1998",
    month        = "jan",
    keywords     = "Optical flow, fluid flow, Meteosat",
    pages        = "256-264",
    journal      = "{IEEE} Transactions on Geoscience and Remote Sensing",
    volume       = "36",
    editor       = "",
    number       = "1",
    publisher    = "",
    url          = "http://www2.compute.dtu.dk/pubdb/pubs/219-full.html",
    abstract     = "The estimation of flow fields from time sequences of satellite imagery
has a number of important applications. For visualisation of cloud or
sea ice movements in sequences of crude temporal sampling a satisfactory
non-blurred temporal interpolation can be performed only when the flow
field or an estimate there-of is known. Estimated flow fields in weather
satellite imagery might also be used on an operational
basis as inputs to short-term weather prediction. In this article we
describe a method for the estimation of dense
flow fields. Local measurements of motion are obtained by analysis of
the local energy distribution, which is sampled using a set of {3-D} 
spatio-temporal filters. The estimated local energy distribution also
allows us to compute a confidence measure of the estimated local normal flow.
The algorithm furthermore utilises Markovian random fields in order to
integrate the local estimates of normal flows into a dense flow field using
measures of spatial smoothness. To obtain smoothness we will
constrain first order derivatives of the flow field.
The performance of the algorithm is illustrated by the estimation of
the flow fields corresponding to a sequence of Meteosat thermal images.
The estimated flow fields are used in a temporal interpolation
scheme."
}