How to check if an image has been doctored

There is a Mobile Optimized version of this page (AMP). Open Mobile Version.

I’ve just come across this neat tool on the internet which will analyse a photo for alterations, providing an ELA image you can see areas that may have been altered highlighted in a lighter color.

Below is an example image taken directly from the website, on the left is the original image and then on the right is the ELA image, areas of suggested modification are shown in bright colours. So from this image there’s some suggestion that the lips, eyes and various parts of the woman's shirt have been modified.

Normal doctored image

Doctored image

Analysis results

Analysis results

How it works?

Here is the explanation given from the website:

“Error level analysis (ELA) works by intentionally resaving the image at a known error rate, such as 95%, and then computing the difference between the images. If there is virtually no change, then the cell has reached its local minima for error at that quality level. However, if there is a large amount of change, then the pixels are not at their local minima and are effectively original.”

-Neal Krawetz, Ph.D. http://www.hackerfactor.com

This implementation makes use of the Python Image Library, and libjpeg (v6.2.0-882.2). ELA is carried out at 95%. Resulting ELA images have had their brightness enhanced to further separate differences out.

The website

If you would like to try this tool out for yourself then waste no time and visit the Image Forensics Website.

Author: Dean Williams

I'm a Web Developer, Graphics Designer and Gamer, this is my personal site which provides PHP programming advice, hints and tips

Post Tags:
, , ,