Text recognition method and device

TL;DR: In this paper, a text recognition method consisting of obtaining, by the text recognition device, a to-be-recognized image containinga target text; determining a feature text region from the to-recognised image according to a preset text recognition network model and preset feature words; extracting text information from the feature text regions according to trained text extraction network model.

Abstract: The invention discloses a text recognition method and device and a storage medium The text recognition method comprises: obtaining, by the text recognition device, a to-be-recognized image containinga target text; determining a feature text region from the to-be-recognized image according to a preset text recognition network model and preset feature words; extracting text information from the feature text region according to a trained text extraction network model, wherein the text extraction network model is composed of four CNN modules, an RNN module and a CTC module; and finally determining a target text according to the text information According to the scheme, the text region related to a feature word is extracted (cut) from a to-be-identified image, the text information is extracted from the extracted feature text region according to a text extraction network model, and then the target text is identified from the text information, so, according to the scheme, the target text can be automatically recognized from the image, only the text information of the text region related to the feature words is extracted, the interference degree of irrelevant texts can be reduced, and the recognition accuracy is high