Method for identifying unsuitable content in colour digital image

Abstract

The invention belongs to the digital image processing technical field, in particular to a method for identifying content in colour digital image. The invention includes an image copy detection algorithm and a pornography identification algorithm. The image copy detection algorithm adopts cascaded method, and firstly comparison of image second order entropy is used for reducing the range in image library. Then SSIM algorithm is used for obtaining similarity in the reduced range. The pornography identification algorithm includes four weak graders, a skin region proportion grader, a face quantity grader, an invariant moment grader and a skin region distribution grader concretely; the four graders output four confidence coefficients, so as to form a four-dimensional characteristic vector, and the vector is input into a strong grader, thus obtaining final result. The higher the result of the strong grader is, the bigger the probability that the content is pornographic is. The system and method has wide application value in the Internet management and control field.

Description

technical field [0001] The invention belongs to the technical field of digital image processing, specifically relates to a color image content analysis and filtering method, and further relates to a digital image copy detection and a human skin segmentation method, the system can be used for unhealthy, violent, reactionary Internet Image detection and filtering. Prevent the spread of unhealthy content and protect the physical and mental health of young people. Background technique [0002] With the increasing use of the Internet, all kinds of bad information (pornography, reactionary, violence, etc.) on the Internet are increasingly interfering with the normal order of the Internet. Due to the lack of effective monitoring of information release by traditional media, a large amount of information that should have been strictly regulated has flooded. How to effectively control the dissemination of these information and ensure the security of the content of the network has be...

AI Technical Summary

Problems solved by technology

[0003] Most of the existing network filtering systems are based on the filtering of URL addresses. This technology simply directly shields the network users from the bad content website URL address database (commonly known as "blacklist") set in the firewall of the network operator. Websites, there are few commercial systems on the market at home and abroad that directly filter the content of online media (images and videos)

Although the URL-based blocking technology is simple and efficient, it has serious limitations: because network operators cannot update the blacklist in time, this will cause many new pornographic websites to slip through the net; at the same time, not all webpages under some domain names The content is unhealthy, which in turn will lead to some normal content being blocked innocently

[0009] In addition, under the premise that the skin segmentation technology is not complete, simply relying on the size of the skin area to judge whether the picture is a pornographic picture will inevitably lead to serious misjudgment

Images