Friday, May 15, 2009

Visual Literacy - First Steps



Here's how I would approach the challenge I posted in the previous blog. The two images, both showing sharks and humans interacting, could be real or fake. One approach is to examine the photos for evidence of 1) tampering and 2) clues for further investigation. A second approach is to search for commentary information about the photos in question.

If the photos are small, it may be difficult to detect any suspicious alterations. In this case, it may be possible to locate larger images by describing what you see. A Google Images search is more powerful than it used to be. A few years ago, too many keywords would bring it to its knees. But now it is possible to use more terms in a query and get results. (That being said, it's always a good idea to use as few keywords as possible since matching each word reduces the likelihood of getting the desired result.)

An image search for the shark and kayaker could simply be shark kayak. This indeed does fetch a larger image--and several related images which could come in handy. A query for the photo of the scuba diver hitching a ride on the shark's fin could simply be shark diver hitch. This query returns a different image of a diver hitching a ride on a shark, which adds some credibility to the claim that a diver might be able to do this. The query shark diver dorsal ride retrieves a larger version of the image shown above.

A closer examination of the shark-kayak photo reveals some irregularities. First, there is a smooth water reflection of the kayaker despite the fact that the surface of the water elsewhere is non-reflective. This suggests that the kayak was pasted into the scene. The photo of the shark could be real. The words on the photo are also helpful: "sitting in a 3.8 meter sea kayak, etc." A Google fact-check of this phrase returns a wealth of pages discussing the photo. The first one of these is: Scary Shark: Real or Not?

Readers' comments are mixed: some think it's real, others don't. This is a good way to use Web 2.0 information to see what the 'crowd' thinks. Taken as a whole, the crowd is usually correct. Of interest here is that a different photo of a kayaker and shark is referenced. This alternate view causes some to doubt whether the photo is actually a fabrication. The reflection is about the same, but now the view of the kayak is close-up and the kayaker is looking a different direction. Now the investigative problem includes the question: were two different photos of the same content faked? Is that likely?

Reader's comments are about 50-50 whether this is real or fake. One of the readers links to snopes.com which claims the photo is REAL. According to snopes it was taken from this source: http://www.whitesharktrust.org/pages/mediaarticle/media25.html. You can see the first appearance of the photo there: a two-page spread in a magazine, hence the line down the middle where the two pages don't exactly match up.

Looking at other returns in Google for the photo reveal a photographer: The author of the photograph is Thomas P. Peschak, a marine biologist at the University of Cape Town’s Marine Biology Research Institute. That name is another great investigative clue. Here's his webpage.

This is what the photographer says about the photo: "When this photograph was first published in Africa Geographic, BBC Wildlife and later in Paris Match and the Daily Mail (London) it resulted in a flurry of e-mails, phone calls and letters from around the world asking if the image was a fake. Following publication no less than 50 websites were hosting colorful discussions about the authenticity of this photograph. The image became probably the most talked about of shark photograph ever and has become my undisputed best seller." [Source]

The other photo--of the diver hitching a ride on a shark's fin--is probably real as well since there are other types of photos of divers hanging onto sharks--too many to all be faked. A similar process could be used to track down references and even who took the picture.

Surprised? Disagree?

No comments: