A novel biologically motivated face recognition algorithm based on polar frequency is presented. Polar frequency descriptors are extracted from face images by Fourier-Bessel transform (FBT). Most of the current face recognition algorithms are based on feature extraction from a Cartesian perspective, typical to most analog and digital imaging systems. The primate visual system, on the other hand, is known to process visual stimuli logarithmically. An alternative representation of an image in the polar frequency domain is the two-dimensional Fourier-Bessel Transform. This transform found several applications in analyzing patterns in a circular domain, but was seldom exploited for image recognition. These results indicate the high informative value of the polar frequency content of face images in relation to recognition and verification tasks, and that the Cartesian frequency content can complement information about the subjects’ identity, but possibly only when the images are not pre-normalized.
Yossi Zana and Roberto M. Cesar Jr, "Face recognition based on polar frequency features", ACM Transactions on Applied Perception (TAP), Volume 3 Issue 1 (2006), pages 62-82.
Index terms: face recognition, fourier coefficients, polar frequency, fourier-bessel transform, fbt, discrete fourier transform, feature evaluation and selection, human perception.
 | Figure 1. 2D Bessel function. | ||
A simple and effective source code for Face Recognition Based on Polar Frequency Features. All tests were performed on AT&T database | |||
Posts
0 comments:
Post a Comment