A text recognition method and device based on tesseract engine

Abstract

The invention discloses a character recognition method based on a tesseract engine and a corresponding device thereof. A cloud server is used to upgrade a local character library to correct the recognition result of the tesseract engine, thereby improving the accuracy of the tesseract engine in recognizing characters in an image. The character recognition device of the present invention includes a server, a tesseract engine module, a cloud server and a local character library. When upgrading the local text library, the server uses the text recognized by the cloud server to correct the recognition result of the tesseract engine module, and supplements the text that the tesseract engine module cannot correctly recognize into the local text library. In this way, when performing character recognition, the recognition accuracy can be improved by querying the local character library. The present invention uses the upgraded local character library to correct the recognition result of the tesseract engine module, which can achieve the same accuracy as directly using the cloud server for character recognition, and can also shorten the time for character recognition calculations, and is suitable for daily learning and image recognition at work. into words.

Description

technical field [0001] The invention relates to an image recognition method, which belongs to the technical field of OCR character recognition (Optical Character Recognition, optical character recognition). Background technique [0002] OCR text recognition refers to the process of electronic devices (such as scanners or digital cameras) checking characters printed on paper, determining their shapes by detecting dark and light patterns, and then using character recognition methods to translate the shapes into computer text; that is, for Printed characters, using optical methods to convert the text in the paper document into a black and white dot matrix image file, and converting the text in the image into a text format through the recognition software, which can be further edited and processed by the word processing software. Optical character recognition (OCR) is the process of converting images of printed text into machine-encoded text. It is widely used to convert data r...

AI Technical Summary

Problems solved by technology

Therefore, it cannot meet the needs of current users to recognize the printed text

[0005] The current image and text recognition technology is either very expensive or has a low recognition rate, which cannot meet the needs of current users for text recognition.

Images