关键词: 混合特征提取 小波变换 复小波变换 OCR 人工神经网络

Abstract:

The paper puts forth the technique for extracting the 2-D wavelet features and the 2-D complex wavelet features based on the Kirsch edge enhancement.  The two types of hybrid features are congregated by combining them with the geometrical features for the recognition of handwritten numerals. In additio n, the merits of the proposed wavelet feature extraction methods are discussed. Experiments based on handwritten numeral recognition and verification sh ow that the two hybrid feature sets can achieve high recognition and verification performance.

Key words: （hybrid feature extraction, wavelet transform, complex wavelet transform, OCR;artificial neural netwo rk）