Computer-based offline handwritten Uyghur word recognition method based on grapheme segmentation

Abstract

A computer-based off-line handwritten Uighur word recognition method based on grapheme segmentation belongs to the word processing technology of character pattern recognition. The steps are as follows: firstly, according to the Uyghur rules and morphological structure, a Uyghur word grapheme database is established, including the main body, additional and Click three types of graphemes; then, obtain three grapheme sequences by over-segmenting the word image, and design different feature extraction and classifiers for each type of grapheme; finally, construct graphemes, connected segments and words through the Bayesian network The layered matching model among them, reasoning and calculating the recognition confidence from grapheme features to word categories, and fusing grapheme recognition information and word formation prior information to obtain word recognition results. Utilizing the off-line handwritten Uyghur word recognition method of the present invention, unconstrained, natural and fluently written Uyghur words can be recognized robustly, and the training category required by the algorithm is fixed, and the algorithm has strong expansibility.

Description

technical field [0001] The invention belongs to the word processing technology of character pattern recognition in pattern recognition, specifically belongs to the field of off-line handwritten character recognition, and is used for recognizing off-line handwritten Uighur word images. Background technique [0002] The Uyghur script belongs to the West-Hungarian branch of the Turkic language family of the Altaic language family. It is the language of the Uyghur minority, an important ethnic minority in my country. The processing and recognition of Uyghur is beneficial to the development of information and technology in ethnic areas. The modern Uyghur language is composed of 32 letters. According to the position in the word, each letter can be written in the form of front connection, double connection, back connection, independent writing, etc., and has evolved into a total of 128 characters. Uighur characters have a long history, and the deformation of handwritten characters ...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the problem that in the existing Uyghur word recognition technology, the segmentation recognition strategy is prone to character over-segmentation errors, and the algorithm scalability of the whole word recognition strategy is poor, and to provide a computer-based detachment method based on grapheme segmentation. Recognition method of handwritten Uyghur words

Images