Advertisement harassment fax image detection system and method based on convolutional neural network

Abstract

The present invention provides a convolutional neural network-based advertising harassment fax image detection system and method, including a keyword area extraction module, the keyword area extraction module is used to determine the keyword suspicious area of ​​the fax image to be detected; neural network Confidence analysis module, the neural network confidence analysis module is connected to the keyword area extraction module, the neural network confidence analysis module is used to identify the words in the keyword suspicious area, and realize the classification of fax images . The present invention extracts the keyword suspicious area through the keyword area extraction module, runs automatically, and has high work efficiency; uses the neural network confidence degree analysis module to identify the text in the keyword suspicious area, realizes the classification and judgment of advertising harassment faxes, and saves time , strong management and control ability, so that the present invention has the characteristics of high work efficiency and strong management and control ability.

Description

technical field [0001] The invention belongs to the technical field of image classification, and in particular relates to a convolutional neural network-based advertising harassment fax image detection system and method. Background technique [0002] With the popularization of Internet technology, the number of text images is increasing day by day, and the automatic processing of text images is an important topic in the field of computer applications. There are many types of text images, and the layout structure is becoming more and more complex. It not only includes text areas with different font sizes, but also often includes elements such as images, tables, and graphics. . Layout analysis is an important link for automatic processing of text images. Layout analysis is to use computers to process text images to determine the physical structure of text images, divide images into areas with different attributes such as text, images, graphics, and tables, and meet different...

AI Technical Summary

Problems solved by technology

[0008] The present invention provides a convolutional neural network-based advertising harassment fax image detection system and method, which can effectively solve the problem of low work efficiency of the existing advertising harassment fax image detection method, and can also solve the existing advertising harassment fax image detection The problem of poor method control ability

Images