Detect fake photos – error level analysis

Image processing was an area I was interested in, but wasn’t interested enough to do some development with it. So while googling around, I found this neat way to detect fake photos. For example, consider a scenario, where I had two screen shots of the same screen, and both of them looked extremely real, except that the content was slightly different. There was no way I could distinguish between the two, unless I was a digital forensic expert.

But, after a looking around I found a neat tool, that which provides a nice analysis of photos. After that, a little bit of intuition and you can identify what part of the photos were faked out.

Here’s how it works.

  1. Upload your images to imgur or any other image hosting service.
  2. Submit the link of the image to
  3. Next you will see a heat map of the image that highlights areas with error.

In the heat map, you can look see that the area that was edited looks a bit out of place. It’s not entirely intuitive, but it’s a neat starting point.

Here’s a great read about the tool: – In the example, notice that there’s a mistake – the fake photo error level analysis is swapped with the real one, compared to the fake and real photos.

Disclaimer: Don’t go making accusations based on what you THINK is fake based on this tool, unless you are an expert. 🙂

    Leave a Reply

    Fill in your details below or click an icon to log in: Logo

    You are commenting using your account. Log Out /  Change )

    Google photo

    You are commenting using your Google account. Log Out /  Change )

    Twitter picture

    You are commenting using your Twitter account. Log Out /  Change )

    Facebook photo

    You are commenting using your Facebook account. Log Out /  Change )

    Connecting to %s