Eccentricity estimator for wide-angle fovea vision sensor

This paper proposes a method for estimating eccentricity that corresponds to an incident angle to a fovea sensor. The proposed method uses Fourier-Mellin Invariant descriptor to estimate rotation, scale, and translation, by taking both geometrical distortion and non-uniform resolution of a space-variant image from the fovea sensor into account. The following 2 points are focused in this paper. One is to use multi-resolution images by Discrete Wavelet Transform for reducing noise caused by foveation properly. Another is to use a variable window function (although the window function is generally used for reducing DFT leakage caused by both ends of a signal) for changing an effective field of view (FOV) in order not to sacrifice high accuracy. The simulation compares the root mean square (RMS) of the foveation noise between uniform and non-uniform resolutions, when a resolution level and a FOV level are changed, respectively. The result shows the proposed method is suitable for the wide-angle space-variant image by the fovea sensor, and, moreover, it does not sacrifice high accuracy in the central FOV. Another simulation result examines a reliable resolution level.