作者: Suneeta V Budihal , R.M. Banakar
关键词: Handwriting 、 Cursive 、 Computer science 、 Handwriting recognition 、 Feature extraction 、 Pattern recognition 、 Computer vision 、 Fourier transform 、 Image segmentation 、 Wavelet 、 Artificial intelligence 、 Wavelet transform
摘要: This paper describes a complete system for recognition of offline cursive handwriting. Preprocessing techniques, which include slant, slope, stroke thickness, segmentation, and normalisation images are described. A new efficient feature extraction method based on Fourier wavelet transform is implemented analyzed. The recognizer starts with features extracted in coarse resolution successive passes renders the same at finer till classification meets acceptance criteria. tested database script proven as an representation compared to other features. Since descriptor rotational invariant, this algorithm works any style