Abstract

An experimental comparison of shape classification methods based on autoregressive modeling and Fourier descriptors of closed contours is carried out. The performance is evaluated using two independent sets of data: images of letters and airplanes. Silhouette contours are extracted from non-occluded 2-D objects rotated, scaled and translated in 3-D space. Several versions of both types of methods are implemented and tested systematically. The comparison clearly shows better performance of Fourier-based methods, especially for images containing noise. Index terms: shape recognition, Fourier descriptors, autoregressive modeling, boundary modeling 1. Introduction Classification of two-dimensional shapes regardless of their position, size, and orientation is an important problem in pattern recognition. Among the areas of applications for invariant 2-D shape recognition are industrial inspection, classification of chromosomes, target recognition, and scene analysis. The methods proposed f...

Citations