An offline handwritten Chinese character recognition method based on dual-tree complex wavelet transform

Abstract

The invention discloses an off-line handwritten Chinese character cognition method based on double-tree complex wavelet transform, which is characterized in that it is carried out according to the following steps: the first step is to analyze the cognition requirements of off-line handwritten Chinese character strokes; the second step is to construct an off-line handwritten Chinese character recognition method; The initial feature model of mobile handwritten Chinese characters; the third step is to construct the candidate subspace of offline handwritten Chinese characters; the fourth step is to design the integrated classifier model of offline handwritten Chinese characters; the fifth step is to recognize the samples of offline handwritten Chinese characters The results are evaluated and the unrecognized off-line handwritten Chinese characters are repeatedly recognized according to the evaluation results. The invention constructs an optimized cognitive model of a specific test sample in the feedback iteration, solves the defect of the fixed cognitive model of an open-loop non-feedback off-line handwritten Chinese character recognition system, and improves the cognitive accuracy of the off-line handwritten Chinese character recognition system.

Description

technical field [0001] The invention belongs to the technical field of Chinese character recognition, and in particular relates to an off-line handwritten Chinese character recognition method based on double-tree complex wavelet transform. Background technique [0002] Chinese character recognition is an important research field of pattern recognition, and it has a wide range of applications in technical fields such as office and teaching automation, automatic recognition of bank notes, and language and text information processing. At present, the research on the feature extraction method of Chinese characters includes academic papers that have made in-depth theoretical analysis, and engineering methods for practical application, such as the invention patent "Chinese Character Recognition Method and Device" (CN102867178A) and the invention patent "Based on Identification and Return Yihua's Handwritten Chinese Character Recognition Method" (CN102831434A). [0003] Chinese In...

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the defects of the existing Chinese character recognition technology, the present invention proposes an off-line handwritten Chinese character recognition method based on dual-tree complex wavelet transform, in order to solve the existing open-loop off-line handwritten Chinese character recognition model when faced with different objects. The same cognitive needs characterize the problem of the feature space of the object to be recognized, so that it can imitate the cognitive characteristics of human beings to adaptively adjust the feature space when facing different cognitive needs, and improve the accuracy of the offline handwritten Chinese character recognition system

Images