How to Tell If an Image Has Been Manipulated or Photoshopped – How-To Geek

Roman Samborskyi/Shutterstock

You cant believe everything you reador see. Social media is rife with manipulated or Photoshopped images. Here are some telltale signs youre looking at an altered image.

Have you ever seen an image that just doesnt look right? Trusting your gut might not be the most scientific approach, but youre probably better at spotting a fake than you realize. If you see an image that sounds alarm bells, you might want to look a bit closer. Youll probably spot some telltale signs that its been manipulated.

Airbrushed images often fall into uncanny valley territory. Even if you have perfect skin, most light sources cast small shadows on fine wrinkles, pores, and other slight imperfections. When these imperfections are digitally removed, so is the appearance of natural lighting.

Professional retouchers often strike a balance between perfection and realism, but amateurs and mobile apps rarely do. Apps, in particular, depend on existing skin tones to determine which parts of a frame to retouch. This often results in a heavy-handed airbrushing effect thats easy to spot.

Sometimes, you might need to look beyond the subject of a photo to see the full picture. This is especially true when it comes to warping, which is when someone uses a tool to grab an area of an image and move, shrink, or enlarge it.

Look for straight lines in the background and see whether they conform to the laws of physics. For example, if someone is sharing an image of their bulging biceps, and a row of tiles in the background is unnaturally warped near said bicep, that photo has been edited to accentuate muscle growth.

This same technique is often used to exaggerate weight loss or the effects of slimming clothing.

Cloning is a basic Photoshop technique that involves duplicating part of an image. Its often used to remove minor blemishes from skin by cloning another section in its place. This also eliminates telltale signs of airbrushing.

This technique is also used in other ways. The object thats duplicated could be a section of a crowd, a tree, or even stars in the night sky. Its an effective way of making a landscape photo pop by dropping in a few more colorful flowers. You can also make a football stadium or event look a lot more crowded than it actually is.

The giveaway in this instance is recognizable patterns appearing in the image. Look for unique aspects in a prominent detail, and then see if you can spot that detail in other parts of the image. It could be someone wearing a unique hat in a crowd, a particular pattern of stars (or constellation), or a tree with the same lighting that appears elsewhere in the image.

This will only apply to the very worst image manipulations, but dont forget to look for a shadow. Its a rookie mistake, but one people still make. Sometimes, an object in an image wont cast a shadow at all.

All objects in a scene should cast a shadow. Plus, if you take a group photo at 5 p.m., you expect the setting sun to cast a longer shadow than an image shot at midday. This can be harder to spot in artificially lit scenes. However, if you see the sun, you should make sure the length and angle of shadows match.

Also, look at how shadows are cast on each subject. If you have a textured object, like a rock, the shadows should look very similar to other textured objects in the image.

When an image has gone through the cycle of being shared, saved, and reuploaded to social media a few times, youll often see compression artifacts. You might spot some unsightly fuzzy sections and colors fringing on hard edges. If an image has been touched up, similar unsightly artifacts often appear right along the edge of the edit.

This is even easier to spot when combined with unusually smooth or solid areas. For example, you might see this if someone attempted to remove text from a white object by painting over it with a white paintbrush. JPEG artifacts often stick to the edge of a painted area like glue.

Any unusually smooth areas with unnaturally solid colors should ring alarm bells, even on high-quality JPEGs.

EXIF data is metadata stored along with a photo when its taken. This includes information like which camera was used, the focal length, aperture, shutter speed, ISO, and so on. Location data in the form of real-world coordinates are also often stored in a photo.

To understand EXIF data, you have to understand a bit more about photography. If the image youre looking at was shot with a very shallow depth of field (like f/1.8), youd expect a very blurry background. A slow shutter speed means any moving objects will be blurred. A long focal length (like 300mm) should compress the background and make a flatter image with a reduced depth of field.

If these parameters (and any others) dont match the image you see, its possible the image has been manipulated. Similarly, EXIF data can contradict a story. For example, say two images were shot close together to portray an extended period of time. If youre lucky enough to have access to thegeolocation data, jump on Google Maps and check out the location using Street View or satellite imagery.

Keep in mind that any editing tools that have been used on a photo, including Photoshop or GIMP, will also be listed under EXIF data. However, this doesnt necessarily mean an image has been manipulated to deceive. There are many legitimate reasons photographers use photo-editing tools, like making minor touchups or for batch edits.

In addition to zooming in and pixel-peeping to spot obvious signs of image editing, there are tools that can help you spot a fake, as well. The most basic of these is a website called Image Edited? that judges whether an image has been retouched.

Image Edited? uses most of the techniques weve covered above to check and report whether any inconsistencies were found. The tool examines EXIF data for inconsistencies in areas like camera models and color spaces. It also looks for JPEG artifacts, oversaturation, patterns that suggest parts of an image have been cloned, and mismatches in directional light.

We tested an obviously manipulated image and Image Edited? reported the image had probably been manipulated because pixels only match software editors.

FotoForensics is similar to Image Edited?, except it leaves the analysis up to you. Instead of making a decision for you, the website produces an Error Level Analysis (ELA) visualization. This can highlight potentially Photoshopped elements you might not catch with the naked eye.

According to the ELA tutorial, you should look around the picture and identify the different high-contrast edges, low-contrast edges, surfaces, and textures. Compare those areas with the ELA results. If there are significant differences, then it identifies suspicious areas that may have been digitally altered.

The best way to get the most out of FotoForensics is to comb through the examples given to learn exactly what to look for. We tried this one with a manipulated photo of a crashed truck with good results. The edited parts of the image clearly contrasted with the rest of the image (see above).

When all else fails, why not search for it? Google Image Search allows you to perform a reverse image search to find other instances of the same image online, as well as images that look similar.This should help you find websites that clearly state the image is a fake, or you might even find the original, unedited version.

You can also search for information about a questionable image on fact-checking websites. For example, lets say theres an image that claims to show little green aliens on the streets of New York City. You could search little green aliens new york to find analyses of the photo, and youd likely find fact-checking articles explaining that those little green aliens arent real.

Thats an extreme example, but the same technique applies to other suspicious or controversial images floating around the web. Do a quick search and a bit of research before you believe what someone claims to be showing.

Photoshopped images are nothing new. Theyve been around and reshared since the birth of the internet. Many people have fallen victim to them in the past. And, as increasingly sophisticated techniques become more accessible, many will fall for them again in the future.

However, now you know what to look for, so youll be better equipped to analyze an image for signs of tampering.

If want to learn how to spot video fakes (or deepfakes),as well, check out this article next!

RELATED: What Is a Deepfake, and Should I Be Concerned?

Visit link:

How to Tell If an Image Has Been Manipulated or Photoshopped - How-To Geek