Long Weibo image identification method and device

Abstract

The invention provides a long Weibi image identification method and device. The long Weibo image identification method comprises steps of obtaining a Weibo image to be identified, converting the Weibo image to be identified to a gray scale image, performing image morphology processing on the gray scale image, wherein the image morphology processing comprises binary processing and erosion and expansion processing, performing character identification on an image which goes through the image morphology processing, and determining the Weibo image to be identified is a long Weibo image when a number of lines of identified characters is greater than a threshold of a preset number of lines. On the basis of the image processing on the Weibo image to be identified and identified processing of effective character lines, the long Weibo image identification method and device can accurately and efficiently identify whether the Weibo image is a long Weibo image. As a result, the long Weibo image identification method and device make data analysis based on long Weibo image identification result more pertinent, information processing redundancy is low and data analysis processing efficiency is high.

Description

technical field [0001] The invention belongs to the field of information processing, and in particular relates to a long microblog picture recognition method and device. Background technique [0002] With the continuous development of social network, its role in people's daily life has become more and more significant, and Weibo, as an important social network medium, has been paid more and more attention by enterprises and government propaganda departments. The evaluation and dissemination track can provide important reference for decision makers. [0003] People can publish their own text comments on an event on Weibo, and they can also publish data information in various forms such as a certain video screen and pictures they have taken. Restricted by the limitation of microblogs on the number of text words, long microblogs (called long microblog pictures) that publish text information in the form of pictures have become a relatively common way of carrying information in ...

AI Technical Summary

Problems solved by technology

[0004] A long Weibo picture is essentially a picture. The first problem to analyze the text data content of a long Weibo picture is: the number of pictures in Weibo is very large, and the proportion of long Weibo pictures is actually not high. , if all the pictures are used such as optical character recognition (Optical Character Recognition, referred to as OCR) technology for text data content recognition, and then data analysis, will lead to very low processing efficiency

Images