In a world with Photoshop or photo editing software at nearly everyone’s fingertips, it can be hard to know if that stunning image making the rounds is real or even recent. Also, it can be hard to track down the original source of photos or images when they pop up on Pinterest with a link to a Tumblr that links to another Tumblr that links to a blog that doesn’t cite the source.
While there’s no fool proof way to find the original, there are a few ways to track down other copies of the image and potentially the original source. One of the easiest places to start is with a reverse image search.
It’s probably a good idea for journalists to plug any images they share into these sites before passing it along or repinning it with credit to the wrong source. Why use it? Last week in the wake of Super Storm Sandy, one of the most shared photos I saw pass around social media was of soldiers standing in a downpour guarding the Tomb of the Unknown Soldier. It was a real shot — from a few months before, not from Sandy as it was being purported to be. (Sandy spawned so many fake images, several places started tracking the real from the fake.) Reverse image searches also could help you find other similar photos of local landmarks that people have taken over the years if you search by one you have.
It’s just another tool in the toolbox and a useful trick when it works. There are a few image search options out there, so if you want to find more just search in your favorite search engine for “reverse image search” and see what comes up. The two I’ll discuss are probably the most well known, but feel free to share more ideas in the comments or links to this post.
This search engine’s sole function is to search for exact match images, including where it came from and how it’s being used, higher resolution images and also modified versions.
I searched for an image I pinned a year ago because it made me laugh (it’s a sign bashing another signmaker for using Comic Sans). I’ve seen it around the web for awhile but the link on Pinterest takes me to a Tumblr log-in page. TinEye turned up 57 results, which I can sort by best match, most changed (for example if someone crops, photoshops, etc.) and biggest image (if I wanted a higher resolution version).
Google Images Search
You’ve probably used Google Image Search to track down images of places, people or things. But did you know you can search by an existing image instead of a name or word? Go to the Google Images search page. Instead of typing in a search term, click on the little camera icon. It will pop up and give you the option of either searching by an image at a certain URL or uploading an image to search by.
Both the URL and uploaded file work the same, basically scanning the web for images that are similar. Unfortunately, this doesn’t always turn up the original in the top results and sometimes the results for “visually similar” images are far far far off-base, but it at least can usually give you a feeler on whether it’s unique. I find it’s useful to skip the search results, and go to the image results with it set it to “more sizes” instead of “visually similar” (the more sizes options means it’s searching for different resolutions but exact copies of your image). You can use the search tools there to sort by the date range. If you get a bunch of results to wade through, you can set the date range to a specific period, so for example you could search the Sandy images with it set to before the storm to see if there are any hits.
In the case of the comic sans sign, Google turned up 256,000 results (evidently this has been reposted a lot), but the earliest references I saw were around July 2010. So I set the date rate to Aug 1, 2010. The image had only been shared a few places by then. The main original web source I can find (correct me if I’m wrong) seems to be Passive Aggressive Notes on July 19, 2010.
Note: This is the third in a series of posts I’ll be sharing on some “basic” web reporting tools, tips and tech skills that journalists new to digital tools may find useful but may not know about or may be embarrassed to ask about. Often, we cover the latest tools and trends on this blog, but for new journalists or those just getting comfortable with using the web or data as a reporting tool, these will hopefully give you a good introduction to build from. If you have something you want covered or an idea of something you think we should explain (for example, that question you’ve had to ask colleagues about — or answer questions about — a dozen times already) please send me a note @meranduh or firstname.lastname@example.org or add your idea in the comments below. If you want to track these as they’re added, they’ll be under the tag “basics.”