This paper describes a new technique of gray-scale characterrecognition that offers both noise-tolerance and affine-invariance. Thekey ideas are twofold. First is the use of normalized cross-correlationto realize noise-tolerance. Second is the application of global affinetransformation (GAT) to the input image so as to achieveaffine-invariant correlation with the target image. In particular,optimal GAT is efficiently determined by the successive iterationmethod. We demonstrate the high matching ability of the proposed methodusing gray-scale images of numerals subjected to random Gaussian noiseand a wide range of affine transformation. The achieved recognition rateof 92.1% against rotation within 30 degrees, scale change within 30%,and translation within 20% of the character width is sufficiently highcompared to the 42.0% offered by simple correlation
展开▼
