Low-pixel-density text recognition method

Abstract

The invention belongs to the technical field of image recognition, and relates to a low-pixel-density text recognition method. The method comprises the specific steps that: image information collection is conducted on a to-be-recognized low-pixel-density text, the collected image information is input and preprocessed, and preliminary positioning is conducted; accurate positioning is carried out again; image information area identification is performed according to the obtained accurate positioning result, the identified result is classified, and the classified data are respectively input intoa special symbol identification model and a common text identification model for identification; and the identified results are summarized and structurally output. The beneficial effects of the invention are that: the method is suitable for large-scale production, the character area containing the special symbol in the low-pixel-density text is judged, the special symbol is recognized by using thetrained model, so that the recognition accuracy of the special symbol is improved, the recognition process is carried out in multiple steps, the involved detection and re-recognition networks are small, and the recognition efficiency can be effectively improved.

Description

technical field [0001] The invention belongs to the technical field of image recognition, and in particular relates to a method for recognizing text with low pixel density. Background technique [0002] At present, a large number of paper documents, such as letters of credit, bills of exchange, invoices, contracts, and transportation documents, will be generated in transactions, trade, and business transactions between enterprises. These paper documents often need to be entered into the corresponding business system, which requires a lot of manual and repetitive labor, and is inefficient and error-prone. [0003] OCR (Optical Character Recognition) technology is a technology that converts text in an image into text through image recognition. Therefore, in the scenario of processing various documents and bills, OCR technology can effectively solve the problems caused by manual entry. [0004] In common documents and bills, there are a large number of special symbols, such a...

AI Technical Summary

Problems solved by technology

[0004] In common documents and bills, there are a large number of special symbols, such as radio boxes, check boxes, underlines, and footnotes of clause constraints, etc., but the OCR recognition systems currently on the market cannot effectively process and recognize them

Due to the low pixel density ratio in these areas, the recall rate of commonly used deep learning detection methods is relatively low, which is prone to missed detection. Difficult to achieve better recognition effect

Images