Multidimensional filtering for movement analysis in image sequences

Movement analysis of visual scenes can be performed by means of linear multidimensional filters, that have already been analyzed in the past. While soundness of results was convincing, interest in those systems declined due to the reduced computational abilities of contemporary computers. Modern hardware allows realization of velocity filters, if the n-D system is carefully adapted to the analysis problem. Fundamental principles of visual scene analysis by linear multidimensional filters are examined with respect to possible sources of degradation. Extraction of movement information and its practical use is demonstrated for application in a blind spot warning system for an automotive driver assistance system. It should be emphasized that in this paper all filtering operations are done in the frequency domain, i.e. realization of transfer functions is not in the scope of this work.