Character area detection method and device and storage medium

Abstract

The invention discloses a character area detection method and device and a storage medium, and belongs to the field of image processing. The method comprises the steps that based on a preset network model, acquiring a probability graph and a position graph of a to-be-detected target image, wherein the position codes are in a negative correlation relationship with the distance between pixel pointsand the central point of a character area to which the pixel points belong; generating a reference image according to the probability graph and the pixel value of each pixel point in the position graph; combining the mutually communicated target pixel points in the reference image to obtain a communicated region in the reference image, and amplifying the contour of the communicated region to obtain a reference contour; and determining a target contour corresponding to the reference contour in the target image, and determining an area corresponding to the target contour as a character area. Compared with the prior art, the character detection method has the advantages that the character areas to which different characters belong are distinguished, the situation that interconnected characters are divided into the same character area is prevented, the influence of character connection on character area detection is reduced, the character area accuracy is improved, and the detection effectis improved.

Description

technical field [0001] The present invention relates to the field of image processing, in particular to a text area detection method, device and storage medium. Background technique [0002] With the rapid development of image processing technology and the widespread dissemination of images in daily life, text recognition technology has become a commonly used image processing technology. Text recognition can recognize text in images and understand the semantic information of images more deeply. The text area detection is a key step in text recognition. Only by accurately detecting the text area in the image can the text in the text area be accurately recognized. [0003] In the related art, a method of using a network model for text area detection is proposed, see figure 1 , to obtain a network model, which is used to obtain the probability that each pixel in any image belongs to a text. Based on the network model, the probability of each pixel in the target image belongin...

AI Technical Summary

Problems solved by technology

[0004] There may be connected text in the target image. It is difficult to distinguish the connected text with the above scheme, and it is easy to divide these texts into the same text area. The accuracy rate is low and the detection effect is not good.

Images