Character recognition method and character recognition device

Abstract

The application provides a character recognition method and a character recognition device. The character recognition method includes: acquiring an image containing to-be-recognized characters; carrying out area segmentation on the image and acquiring a text block of each area; sequentially inputting each text block into at least two recognition modules at the same time for recognition, wherein each of the recognition modules has a respective confidence degree, and the confidence degrees of the recognition modules are different; taking a recognition result of the recognition module with the high confidence degree as a recognition result of the current text block when recognition results of the recognition modules are different; and integrating recognition results of all the text blocks to obtain a result of recognition for the characters in the image.

Description

technical field [0001] The invention relates to the technical field of character recognition in pattern recognition, in particular to a character recognition method and a character recognition device. Background technique [0002] Character recognition, mainly to recognize Chinese characters or English characters on images, by taking pictures or intercepting images containing characters, then identifying the character content in the image, and finally outputting the character content in the recognized image into editable electronic text . [0003] At present, there are mainly the following technologies for character recognition: [0004] The first method uses a sliding window to segment the target image, and compares and recognizes the segmented blocks with the sample set. However, handwritten samples are diverse and varied, and the size of the frame is not easy to determine; [0005] The second method collects the user's writing track when the user is handwriting input, a...

AI Technical Summary

Problems solved by technology

[0004] The first method uses a sliding window to segment the target image, and compares and recognizes the segmented blocks with the sample set. However, handwritten samples are diverse and varied, and the size of the frame is not easy to determine;

[0005] The second method collects the user's writing track when the user is handwriting input, and recognizes handwritten characters by judging the track trend, but this method can only recognize the track written by the user on the spot, and cannot recognize the handwritten character image that lost the user's writing track;

[0006] The third is to recognize the characters in the image scanned by the scanner containing handwritten characters through the traditional optical character recognition technology. However, the clarity of the image captured by the smart phone is much worse than that of the scanner. Recognition of handwritten characters on a poorly defined image

[0007] All of the above methods for character recognition have deficiencies, and cannot quickly and accurately recognize characters in a large character image into electronic text at one time

Images