Character recognition method and device

Abstract

The invention provides a character recognition method and device. The method comprises the steps of collecting a character image including a character to be recognized and conducting character image preprocessing; uploading an image obtained after the preprocessing to a cloud and conducting unstructured storage, meanwhile, adopting an OCR model to conduct OCR, after the recognition, storing the recognition result and character features under the corresponding character in a character feature pool, and adding the multi-aspect features of the same character; retraining the OCR model by means of the updated character feature pool. The device comprises a preprocessing module, a recognition module, a storage module and a training module. By means of the method and device, the adaptation to recognizing character images under different application contexts and at different shooting angles in the practical application process can be improved.

Description

technical field [0001] The invention relates to the field of computer application technology, in particular to a character recognition method and device. Background technique [0002] Digital pictures include a variety of information, for example, the ID picture contains the bearer's personal information, and the digital image contains landscapes, portraits, etc. In many scenarios, the computer needs to be able to read the text information directly, which requires OCR (optical character recognition) optical character recognition. For printed characters, the text in the paper document is converted into black and white dot matrix in an optical way. Image files, and the text in the image is processed into a text format through the recognition software, and how to accurately process the text through the recognition software is an important technical issue for OCR text recognition at present. [0003] With the full development of my country's informatization construction, OCR ch...

AI Technical Summary

Problems solved by technology

[0004] The fonts in the photographed or scanned pictures, such as a picture containing a certificate, can obtain information about the person, which provides a good data source for tasks such as person retrieval, personal information database update, and person relevance query. Most of the text contained in the text has different fonts, sizes, and complex shooting orientations. However, the current recognition methods used in the existing technology have low adaptability to texts that occur in the shooting scene or the shooting angle changes, and the recognition efficiency is not ideal.

Images