PetaPixel

Izitru: A Service that Helps You Prove You Didn’t ‘Shop that Photo

hightrust

As digital photographs become easier and easier to create, edit, and share, it’s also becoming easier to doubt the authenticity of photos. There have been quite a few stories in recent days of photojournalists, news organizations, and contest winners throwing their reputations away by using Photoshop to manipulate the truthfulness of photos.

Izitru (pronounced “is it true”) is a new free service that aims to make it easier for honest photographers to prove the authenticity of their images.

The system revolves around a website that will automatically examine and validate uploaded photographs, allowing you to show off a 3rd-party stamp of approval if your integrity is ever challenged. There is also an izitru iOS app that will be appearing in the App Store starting today.

Here’s how it works: first, upload your original, unmodified JPEG photograph to the izitru website. The website then analyzes the image based on 29 dimensions of compatibility six separate forensic tests that are commonly used to test images (devise signature analysis, JPEG structure analysis, double JPEG detection, JPEG coefficient analysis, sensor pattern analysis, and JPEG ghost detection).

After analysis, izitru will create a dedicated page for the photo that displays one of five trust levels: high trust, medium trust, undetermined file history, potential file modification, and no trust. Photographers can share this page in order to “prove” the legitimacy of their image.

We did a quick test on the service by uploading a fake hurricane photo that made the rounds a while back. The service correctly identified it as not being reliable:

sorry

If you ever have an authentic image be marked as manipulated by izitru, the service also offers a challenge feature that can mark the photo for examination by human experts.

izitru is a new product offered by the company Fourandsix (pronounced “forensics”), which also offers a software suite and Photoshop plugin for detecting manipulated images.