How to Detect Fake Photos

Because of the widespread use of digital technology, generating fake pictures has never been easier. Making fake photos of celebrities to impress your friends or doctoring media photos to alter public opinion is just as easy. Fakes are created either by merging two or more photos or altering an existing image. Because image manipulation happens at the pixel level, detection is not as easy as it was before the digital era. Tricky fakes can be exposed by algorithms that detect discrepancies or statistical irregularities at the bit level.

Difficulty:

Challenging

Instructions

Things You'll Need

• Image Editor

1. Look for obvious anomalies in the photo. This can include, objects not in the correct perspective, inconsistent angles of reflection or unexpected discontinuities. Check for inconsistent shadows or features that seem slightly larger or smaller than normal. Any inconsistencies may reveal the photo to be a fake.

2. Launch Paint, the image editing tool that comes with Windows, open the image and zoom into the area you want to scrutinize to detect if the photo is a fake. See if you can find pixilation halos around the outline. If colors or lines do not line up just right, then the photo may be a fake.

3. Submit your photo to a professional image processing lab if the previous steps didn't turn up unusual artifacts but you still suspect the image might have been doctored. The lab will be able to detect if the photo is a fake.

4. Buy your own image analysis software if you have a constant supply of photos to analyze. Image analysis software can be expensive.

Tips & Warnings

• To open Paint in Windows, access Programs -> Accessories -> Paint.

• Many digital cameras display 'pixilation halos' when photographing objects against very bright backgrounds. Before concluding that the object under scrutiny has been added check the outlines of other objects.

Resources

• Outsourcedataprocessing.com offers a image processing service.
• Image processing software can be found at mediacy.com.
• A tutorial on fake photo detection can be found at ph.tn.tudelft.nl.