Invariant neural-network based face detection with Orthogonal Fourier-Mellin Moments

In this paper, we apply a recently developed type of moments, Orthogonal Fourier-Mellin Moments (OFMMs) [7], to the specific problem of fully translation-, scale-, and inplane rotation-invariant detection of human faces in two-dimensional static color images, and we compare their performance with that of the generalized Hu's moments or non-orthogonal Fourier-Mellin moments (FMMs). As compared to the FMMs, the OFMMs have the advantages of non-redundancy of information, robustness with respect to noise and the ability to reconstruct the original object [7]. Color segmentation is first performed in nine different chrominance spaces by use of two human skin chrominance models as described in [10]. For feature extraction in the segmented images, the same number of OFMMs are used as for the FMMs as the input vector to a multilayer perceptron neural network (NN) to distinguish faces from distractors. It is shown that, at least in the specific problem of face detection from segmented images, for the same set of test images, there is no significant advantage over the FMMs in using the OFMMs, and that in practice both types of moments may be used. Possible explanations for such results are presented.