Image recognition method and device, terminal and storage medium

Abstract

The embodiment of the invention discloses an image recognition method and device, a terminal and a storage medium. According to the embodiment of the invention, the method comprises steps of obtaininga to-be-identified image, wherein the to-be-identified image is an image of a to-be-identified certificate; performing text area recognition on the to-be-identified image to obtain a text area imageof a target text corresponding to the to-be-identified certificate; determining a text direction of the target text according to the text region image; adjusting the direction of the text area image according to the text direction to obtain an adjusted text area image; and performing text recognition based on the adjusted text region image to obtain text content of the target text. According to the method, the text region image of the target text in the image to be recognized can be extracted, the text direction of the target text is determined, and after the skewed and inverted text region image is corrected automatically by adopting the text direction, the text region image can be used for text recognition, so that the text recognition accuracy is improved. Therefore, the scheme can improve the efficiency of the image recognition method.

Description

technical field [0001] The present invention relates to the field of image processing, in particular to an image recognition method, device, terminal and storage medium. Background technique [0002] At present, computers can recognize the target text of physical documents such as bank cards, ID cards, and membership cards. For example, users can use their smartphones to take pictures of their bank cards. The card number displayed on the face of the card does not need to be manually input by the user, and the smartphone can automatically enter the user's bank card number. [0003] However, once the user cannot clearly and completely take a photo of the physical certificate from the front, or the layout of the physical certificate is different from common physical certificates, the user needs to re-photograph the physical certificate, otherwise it will lead to unrecognizable, wrongly recognized, slow recognition speed, etc. , thus, current image recognition methods are ineff...

AI Technical Summary

Problems solved by technology

[0003] However, once the user cannot clearly and completely take a photo of the physical certificate from the front, or the layout of the physical certificate is different from common physical certificates, the user needs to re-photograph the physical certificate, otherwise it will lead to unrecognizable, wrongly recognized, slow recognition speed, etc. , thus, current image recognition methods are inefficient

Images