Technologies to look for online pornographic images of an abusive nature will be developed as part of an attempt to tackle digital sex crimes, announced at a government-ruling party meeting Tuesday.

As part of the government-led initiative against digital sex crimes, automated systems to detect and flag the images for removal will be developed by state-run research institutes under the Ministry of Science and ICT.

The new tools, which are already under development, will combine machine learning, artificial intelligence and text analysis to scrutinize content shared on peer-to-peer tools such as online hard drives.

The toolkits will extract key visual and sound features to automatically recognize and flag images collected in an abusive way or without the subjects‘ knowledge, such as child pornography, revenge porn and hidden camera footage.

Up to now, this task has been carried out mostly by human content moderators.

Jung Young-gil from the ICT policy department of the Science Ministry told The Korea Herald that the toolkits will help process vast amounts of online content on a real-time basis, with the aim of removing it.

The toolkits under development are expected to be put into use at the Korea Communications Standards Commission to step up its monitoring efforts as early as in 2019.

(mjk625@heraldcorp.com)