Text detection method and device, computer device and storage medium

Abstract

The invention relates to a text detection method and device, a computer device and a storage medium. The method comprises the steps of obtaining a to-be-detected picture; extracting picture features of the to-be-detected picture based on different scales; determining character related information of each character in the to-be-detected picture corresponding to each scale according to the picture characteristics of each scale; and integrating the character related information corresponding to each scale to obtain a character detection result of each character in the to-be-detected picture. According to the method, character detection is realized by extracting picture features from different scales, and the method can adapt to characters of different scales in a natural scene, so that the problem of inaccurate text detection caused by inconsistency of picture sizes and the like is avoided.

Description

technical field [0001] The present application relates to the technical field of image processing, in particular to a text detection method, device, computer equipment and storage medium. Background technique [0002] The text detection method in the natural scene can be used to locate the text information in the image, so that the subsequent text recognition can be performed on the detected text box pictures containing text, and finally the structured information can be obtained. Text detection methods are widely used in image retrieval, video analysis, automatic driving and other fields, and the accuracy of text detection affects the accuracy of subsequent text recognition and information structuring processes, so the development of an effective text detection in natural scenes Algorithms are crucial. [0003] However, in actual natural scenes, there are usually various uncontrollable interference factors, such as image brightness changes, distortion of captured images, d...

AI Technical Summary

Problems solved by technology

[0004] At present, the more commonly used text detection can be realized by artificial intelligence technology, mainly including text detection and character recognition models. However, this method usually requires a high degree of precision. If a character is positioned incorrectly, it will seriously affect subsequent text recognition. And the performance of information structure, but in the real scene due to the limitation and complexity of shooting conditions, there are big differences in the text images shot by different users using different hardware, mainly in different brightness, different sizes, and different angles , There are problems such as occlusions, so it is easy to cause inaccurate text detection results

Images