PhotoDNA is a technology developed by Microsoft and improved by Hany Farid of Dartmouth College that computes hash values of images, video and audio files to identify alike images. PhotoDNA is primarily used in the prevention of child pornography, and works by computing a unique hash that represents the image. This hash is computed such that it is resistant to alterations in the image, including resizing and minor color alterations. It works by converting the image to black and white, re-sizing it, breaking it into a grid, and looking at intensity gradients or edges.
It is used with Microsoft's own services Bing and OneDrive, as well as by Google Gmail, Twitter, Facebook and the National Center for Missing & Exploited Children, to whom Microsoft donated the technology.
Microsoft donated the PhotoDNA technology to Project Vic, managed and supported by the International Centre for Missing & Exploited Children (ICMEC). Project Vic is an image and video hash-sharing initiative that streamlines investigative workflows and narrows the focus of child pornography law enforcement investigations by filtering the material investigators find on offenders’ computers. Project Vic uses the technology to create a "fingerprint" that can be used to uniquely identify an individual photo. Using robust hash sets, the technology allows law enforcement to determine which images retrieved have already been identified, and are part of the Project's database of millions of digital hashes of child porn, enabling detectives to focus on those that are new children waiting to be located and recovered. The technology also assists online service providers, by helping them detect child sexual abuse images shared on their sites, and block their continued dissemination.
In December 2014, Microsoft also made PhotoDNA available to qualified organizations as a free cloud service through the Azure Marketplace.
In late 2015, Farid completed improvements to the software which now make it capable of analyzing video and audio files besides still images. In 2016, Hany Farid proposed that the technology could be used to stem the spread of terror-related imagery, but there was little interest shown initially by social media companies. In December 2016, Facebook, Twitter, Google and Microsoft announced plans to use PhotoDNA to tackle extremist content such as terrorist recruitment videos or violent terrorist imagery.