Interpol to use Microsoft PhotoDNA to help ID child abuse victims
By Digital News Asia April 23, 2015
- Can identify images even if they have been modified or edited
- Integration to database for specialised investigators from 46 countries, Europol
INTERPOL will be integrating Microsoft Corp’s PhotoDNA technology into its International Child Sexual Exploitation (ICSE) database as part of the world police body’s efforts to identify victims of child abuse worldwide.
PhotoDNA calculates the distinct characteristics of a digital image, creating a unique signature, also known as a ‘hash.’
This can then be compared with the hashes of other images to efficiently and reliably find matching signatures, Microsoft said in a statement.
Where the PhotoDNA tool differs from other hash algorithms is that it finds images even if they have been changed, making it possible to find many more matches than with classic hash, the company added.
The PhotoDNA signatures will be produced from the ICSE database, along with other ‘hashing’ algorithms producing specific signatures, against which member countries can compare their files for potential matches.
“It is essential for all child abuse and exploitation material to be analysed as thoroughly and efficiently as possible to help identify and rescue victims as quickly as possible,” said Mick Moran, head of Interpol’s Human Trafficking and Child Exploitation unit.
“What is also essential is eliminating duplication of effort, to avoid officers spending valuable time trying to identify a victim who may have already been saved.
“PhotoDNA, as part of the ICSE database, will help our member countries address these issues,” he said in Singapore.
Specialised investigators from 46 countries and Europol currently have direct access to ICSE, with connection to a further 25 countries under way.
“The problem of trading child sexual abuse images, and the re-victimisation for those whose pictures continue to surface online, is very troubling,” said Courtney Gregoire, senior attorney at Microsoft’s Digital Crimes Unit.
“PhotoDNA, which can identify the ‘worst of the worst’ images, even if they have been altered, can help address this problem,” she added.
PhotoDNA will also be available as a signature list for Interpol’s Baseline project which seeks to empower industry and network administrators to recognise, report, and remove child abuse material from their networks.
The project, currently in the pilot phase, allows industry to cross-match image signatures hosted on their network against signatures extracted from some of the worst child abuse materials in ICSE.
Accessible to victim identification specialists through Interpol's secure I-24/7 global police communications system, ICSE also uses sophisticated image comparison software to make connections between victims and places.
By the end of March 2015, the database had assisted in the identification of more than 6,300 child sexual abuse victims and the arrest of more than 3,200 offenders.
How it works
PhotoDNA technology converts images into a common black-and-white format and uniform size, then divides the image into squares and assigns a numerical value that represents the unique shading found within each square.
Together, those numerical values represent the ‘PhotoDNA signature’ or ‘hash’ of an image, which can then be compared against signatures of other images, Microsoft said.
While the technology cannot be used to identify a person or object in an image, nor can it be used to recreate an image, it can be used to find copies of a given image with incredible accuracy and at scale across the 1.8 billion images shared online every day, even when the images themselves have been altered.
For an FAQ on PhotoDNA, click here.
500mil kids going online in emerging Asia, need for more safety
Net censorship: Do the ends justify the means?
ViSenze's mission is to make sense of the visual web
Security researchers identify malware threat to virtual currencies
For more technology news and the latest updates, follow us on Twitter, LinkedIn or Like us on Facebook.
Author Name :
By commenting below, you agree to abide by our ground rules.