Detection of pornographic content in internet images

Evaluation of local descriptors and CNNs for non-adult detection in visual content

“…The first step, pornography detection, has been accomplished making use of skin detection as in Jones and Rehg (1998), or more recently combining different cues such as color, texture and shape also used in Sengamedu et al (2011). Our proposal is devoted to partially solving the aforementioned second stage, describing an approach to classify an individual as adult or not.…”

Incremental Learning for Compressed Pornographic Image Recognition

“…Nowadays, pornographic images are inevitably spread and flooded on the network, which have done great harm to people's physical and mental health, especially for adolescents. Therefore, how to efficiently recognize and filter the pornographic information has become an urgent issue for network information security [3][4][5].…”

“…In recent years, researchers have proposed and explored the pornographic image recognition methods based on image content, such as body structural features [3], skin color features [4] and visual words features [5,6], etc. Many researchers have proven the distribution information of visual words feature of the pornographic images can more efficiently describe the image content than the traditional underlying lowlevel visual feature.…”