Affine-invariant recognition of gray-scale characters using global affine transformation correlation

This paper describes a new, promising technique of gray-scale character rec ognition that offers both noise tolerance and affine-invariance. The key id eas are twofold. First is the use of normalized cross-correlation as a matc hing measure to realize noise tolerance. Second is the application of globa l affine transformation (GAT) to the input image so as to achieve affine-in variant correlation with the target image. In particular, optimal GAT is ef ficiently determined by the successive iteration method using topographic f eatures of gray-scale images as matching constraints. We demonstrate the hi gh matching ability of the proposed GAT correlation method using gray-scale images of numerals subjected to random Gaussian noise and a wide range of affine transformation. Moreover, extensive recognition experiments show tha t the achieved recognition rate of 94.3 percent against rotation within 30 degrees, scale change within 30 percent, and translation within 20 percent of the character width along with random Gaussian noise is sufficiently hig h compared to the 42.8 percent offered by simple correlation.