Invertible Orientation Bundles on 2D Scalar Images

A general approach for multiscale orientation analysis of 2D scalar images is proposed. A scale-dependent orientation bundle (map of the visual space into function of two arguments: position and orientation) is constructed from the local Gaussian-derivatives jet of a scalar image in 2D. It is shown that there exists a class of orientation filters exhibiting an invertible relation between the orientation bundle and the original image in space domain. This invertible transformation is used to regain the original acuity in the spatial domain after analyzing orientation features at any given scale. The approach turns out to be highly effective for the detection of elongated structures.