Skin Detection based Pornography Filtering using Adaptive Back Propagation Neural Network

Abstract

As the internet becomes faster and cheaper, its misuses like pornographic production and consumption has also been increased. Pornography is considered a sensitive issue to discuss openly in our society and this is a neglected one too. Psychological research says that Pornographic and nude images create a negative impact on the viewer's mind. And also watching pornography is a kind of addiction too. At the first stage, such people create distance from their loved ones which leads them to depression and on extreme stages they could be involved in many types of criminal activities. In this article, the Skin Detection based Pornographic Filtering using Adaptive Back Propagation Neural Network (SD-PFT-ABPNN) Technique is presented. The Simulation results of Proposed SD-PFT-ABPNN techniques shown desirable results regarding MMSE and regression as compared to conventional skin detection-based Porn Filtering Techniques using Global Image Enhancement (PFTGIE), Porn Filtering Techniques Without using Global Image Enhancement (PFTWGIE) techniques. When the results were compared, it was seen that the BR algorithm has the highest accuracy rate with 99.70%.