Picture-based font recognition method and device and an electronic device

Abstract

An embodiment of the invention discloses a picture-based font recognition method and device and an electronic device, relates to character recognition technology and can help effectively improve fontrecognition accuracy. The picture-based font recognition method comprises the following steps: wavelet transformation is performed on an image to be recognized so as to obtain a first wavelet featurevector according to a preset wavelet transformation series; feature transformation is performed on the first wavelet feature vector to obtain a first transformed feature vector according to a linear transformation matrix obtained in advance via sample calculation; a decision function value is calculated according to the first transformed feature vector and a feature parameter of a sample font obtained in advance via the sample calculation; according to the decision function value, the font of characters in a picture to be identified is determined. The picture-based font recognition method anddevice and the electronic device disclosed in the invention are suitable for performing font recognizing operation on characters contained in pictures.

Description

technical field [0001] The invention relates to character recognition technology, in particular to a picture-based font recognition method, device and electronic equipment. Background technique [0002] Optical Character Recognition (OCR, Optical Character Recognition) or normalized recognition is to use electronic devices such as scanners or digital cameras to check printed characters, determine the shape of characters by detecting dark and bright patterns, and then use character recognition methods to convert the determined characters The technology of translating shapes into computer characters belongs to the field of pattern recognition (PR, Pattern Recognition), and is mainly used in character recognition. [0003] As graphics and text can provide users with more and more exciting information, the application of text in pictures is becoming more and more common. Therefore, how to identify characters and their fonts from graphics and texts so that users can determine whi...

AI Technical Summary

Problems solved by technology

However, the currently applied optical character recognition or normalized recognition is mainly used for pure character recognition. For the fonts of the characters contained in the picture, the fonts of the characters contained in the picture cannot be recognized or the recognized fonts do not match the actual fonts, which makes the accuracy of font recognition relatively low. Low, resulting in lower accuracy in setting characters based on recognized fonts

Images