Text processing method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention provides a text processing method and device, electronic equipment and a storage medium, and relates to the technical field of image processing. The method comprises the following steps: performing character detection on a video frame sequence of a current video, and determining a plurality of text lines; according to the position information of the text lines, the text lines are initially classified, a first set and a second set are obtained, the first set is recognized as line texts, and the second set is recognized as non-line texts; according to the font feature information of the text lines and a preset clustering network, the text lines are clustered to obtain a plurality of clustering results, and the text lines in the same clustering result have the same font; and according to the plurality of clustering results, carrying out secondary classification on the text lines, and determining a final line text of the current video. According to the method, secondary filtering is carried out on the preliminarily classified text lines through the font feature information, non-line texts which are detected by mistake in line texts can be filtered out, line texts which are not recognized can be recalled, and the accuracy is improved.

Description

Technical field [0001] The present invention relates to the field of image processing technology, in particular to a text processing method, apparatus, electronic devices and storage media. Background [0002] In the fields of text search, text recognition, text translation, etc., it is first necessary to locate the target text in the target area, and then the target text can be recognized and translated. At present, the common method is to determine whether the line of text belongs to the target text of the target area by using the position information of the line of text. However, only the position information used for the line of text is used to judge it, which will cause when the non-target text appears in the target area, due to the successful matching of the location information, resulting in the non-target text being mistakenly identified as the target text. There will also be a large error in the position of the target text or a situation where the match fails and is misf...

AI Technical Summary

Problems solved by technology

However, only using the position information of the text line to judge it will cause that when other text lines appear in the position of the line text line, the successful matching of the position information may cause the non-line text line to be misidentified as the line text line

In the same way, there will be a large error in the position of the line text line or it is just at the critical threshold, which will lead to the failure of matching and false filtering

Images