[ad_1]
July 6, 2018
Christian Haschek, a security researcher, has built a system to detect and eliminate illegal images costing less than $ 300 and consuming less energy than a light bulb. The resident researcher in Austria, proposed the solution after discovering an image that showed that the child sexual abuse had been loaded onto their Pictshare image hosting platform. He called the police, who told him to print it and bring it to him. However, it is illegal to have images of child abuse, digitally or on paper. "Uh … it's not what I planned to do," Haschek said.
Haschek used three PI Raspberry, feeding two Intel Movidius sticks, which can be trained to classify images. He also used an open source algorithm to identify the explicit hardware called NSFW (Not Safe for Work), available for free from Yahoo. Christian Haschek, in a few hours, has put himself in the lead to find images that the system could say with 30% or more certainty that probably contains pornography, said that he was unlikely to do not miss anything
16 more illegal images with children on his platform, all he reported to Interpol and erased them. He then contacted a larger picture hosting service, which he declined to mention, and found thousands of other images loaded on his platform through his system. .
Keep Reading: The $ 300 System In The Fight Against Illegal Images – BBC News
<! – Trackbacks of Autodiscovery
Source link
