OCR (Optical Character Recognition) result classification method based on knowledge graph

Abstract

The invention discloses an OCR (Optical Character Recognition) result classification method based on a knowledge graph. The method comprises the following steps: constructing an ontology of an OCR result knowledge graph; constructing a text classification model and a named entity extraction model to form a classifier; and constructing a knowledge graph based on an OCR recognition result and the classifier constructed in the step S2. According to the method, the ontology is constructed according to the text classification information in the specific field, the text classifier is constructed based on the ontology, the classification and key information of the OCR software recognition result are extracted by utilizing the classifier, and the text knowledge graph is constructed, so that the purposes of automatic multi-level classification and key information extraction of the OCR recognition result are achieved. According to the method, automatic multi-level classification and key information extraction of the OCR recognition result can be realized, and the problems that the hierarchical relationship between categories is ignored and the classification refining degree is not high in the existing similar technology, and the problems that only the category information of the text is extracted and stored, other information in the text is ignored and text retrieval is inconvenient are solved.

Description

technical field [0001] The invention belongs to the technical field of image OCR (Optical Character Recognition, optical character recognition), and specifically relates to a method for classifying OCR recognition results based on a knowledge graph. Background technique [0002] OCR (Optical Character Recognition, Optical Character Recognition) technology refers to the use of electronic equipment (such as scanners or digital cameras) to convert paper documents into image files of black and white dot matrix, and determine the characters in the image by detecting dark and bright patterns. shape, and then use character recognition to translate the shape into computer text for further editing and processing by word processing software. OCR software refers to software that uses OCR technology to digitize paper documents, and is widely used in various fields of production and life. [0003] Units and institutions that require the use of OCR technology often have the characteristi...

AI Technical Summary

Problems solved by technology

Both schemes realize the automatic classification of OCR recognition results to a certain extent, but both schemes have shortcomings: the first one ignores the hierarchical relationship between categories, and the degree of refinement of classification is not high

The second is to extract and store the category information of the text, ignoring other information in the text, which is not convenient for text retrieval

Images