The Microsoft PhotoDNA Cloud Service has been launched, which will help the law enforcement agencies and online businesses to identify illegal images, which includes the images showing sexual abuse of children, from enormous photo libraries uploaded on the internet.

Microsoft PhotoDNA was developed by Microsoft in collaboration with Darthmouth College. Since 2009, its on-premise version has been used by various organizations and tech companies including Facebook and Twitter, but to run PhotoDNA, it requires considerable money, time and expertise. It is easy to set up and deploy the new Microsoft PhotoDNA Cloud Service from Microsoft’s Azure platform. Some qualified organizations can use it for free to automatically detect and report the distribution of images showing sexual abuse of children.

How Microsoft PhotoDNA cloud service works?

According to Microsoft 1.8 billion images are uploaded to the internet every day and nearly 720,000 of them are images containing exploitive content. The Microsoft PhotoDNA cloud service works by hashing and converting images into digital values, which are then matched against databases of hashes from known illegal images contributed by trusted sources like the NCMEC. The new service can also identify images that have been modified to escape detection, which means that if an image has been resized or had a mark added to it, it can still be tracked.

David Creemer, Flipboard’s head of Platform Engineering said “The Flipboard community is built on the desire to inspire each other with the things we love. Our community needs to trust that we do everything possible to stop the spread of illegal content, especially images of child sexual abuse.”

The Microsoft PhotoDNA Cloud Service has been recently adopted by Flipboard and Kik. Kik has more than 200 million users and it uses PhotoDNA to detect exploitive profile pics so that they can be removed. The company also reports user’s account to law enforcement agencies that use exploitive images. Microsoft also said that it has improved the algorithm which is used to identify illegal images and is now 1000 times faster than the previous version.