Proofreading method, system and proofreading server of characters in ancient book

Abstract

The invention provides a proofreading method of characters in an ancient book. The method comprises the following steps that: an ancient book image is obtained by a proofreading server, and segmented into a plurality of character image blocks; each character image block is identified; the similarity between each character image block and a standard character is calculated; when the similarity between each character image block and the standard character is less than or equal to a first preset value and greater than a second preset value, the corresponding character image blocks are clustered, and stored to at least one to-be-selected set with the standard characters as an index; the character image blocks in the to-be-selected set and the index corresponding to the to-be-selected set are provided to a client to be judged by a user of the client; and the character image blocks in the to-be-selected set are proofread according to a judgment result of the user. The proofreading method can effectively improve the identifying accuracy of the character corresponding to each character image block in the ancient book image, and ensures the translation quality of the ancient book image. The invention further provides a proofreading system and the proofreading server of the characters in the ancient book.

Description

technical field [0001] The invention relates to the technical field of information sharing, in particular to a proofreading method for Chinese characters in ancient books, a proofreading system for Chinese characters in ancient books and a proofreading server. Background technique [0002] There are so many ancient books in ancient China, but most of them exist in the form of scanned pictures. Due to the complex layout of ancient books and the existence of a large number of handwritten characters, the current text recognition effect of ancient books is not good, and the investment is large. Small. [0003] At present, there are a large number of Chinese books whose copyright has expired, scanned copies of periodicals from the Republic of China period, and some electronic books that have been manually processed, but most of them are of low quality and have a high rate of text errors. Contents of the invention [0004] The present invention aims to solve at least one of the...

AI Technical Summary

Problems solved by technology

[0002] There are so many ancient books in ancient China, but most of them exist in the form of scanned pictures. Due to the complex layout of ancient books and the existence of a large number of handwritten characters, the current text recognition effect of ancient books is not good, and the investment is large. Small

[0003] At present, there are a large number of Chinese books whose copyright has expired, scanned copies of periodicals in the Republic of China, and some electronic books that have been manually processed, but most of them are of low quality and have a high rate of text errors.

Images