Image: Courtesy AntesyDespues.com
-
The Best Science Writing Online 2012
Showcasing more than fifty of the most provocative, original, and significant online essays from 2011, The Best Science Writing Online 2012 will change the way...
Read More »
Scientists in the United States have come up with a tool for automatically analysing digital photographs, making it possible to gauge the extent to which images have been altered or retouched.
Advances in image-manipulation software have made it trivial to radically alter the appearance of models and celebrities in photos, notes Hany Farid, a computer scientist who studies digital forensics and image analysis at Dartmouth College in Hanover, New Hampshire. Farid created the analysis tool with his colleague Eric Kee, also at Dartmouth College. The promotion of unrealistic body images in some advertisements and magazines is thought to have a role in triggering eating disorders, explains Farid, and some countries, including the United Kingdom, France and Norway, are now considering legislation to require digitally altered images to be labelled as such.
The idea is to use the software to generate a scale that can be printed next to published images, say Farid and Kee, so that readers can tell how accurately they represent the originals. The hope is that this will shed light on the culture of 'airbrushing' in the advertising and fashion-magazine industries. The software could also help to deter fraud in scientific images, they say.
However, simply labelling manipulated images is not the solution, says Farid, because this would tar all altered images with the same brush — even those that used legitimate adjustments such as cropping and colour modification. Farid and Kee's solution, published online today in the Proceedings of the National Academy of Sciences USA, is a system that can score on a scale of one to five how much an altered image has strayed from reality.
Compare and contrast
Farid and Kee first compared more than 450 pairs of images before and after manipulation, quantifying their dissimilarity according to eight different statistical parameters. These ignore any global changes, such as cropping, and instead focus on local geometric modifications—for example, by how many pixels the shape of a person has altered—and photometric changes such as smoothing or sharpening.
To combine these parameters into one metric, the researchers asked more than 350 volunteers to compare the same pairs of images, ranking them on a scale of 1 (very similar) to 5 (very different). These ratings were then used to train a machine-learning algorithm to extract a single score from the measured values that would faithfully reflect the perceptual judgement of the volunteers.
The resulting system is able to rate the extent of manipulation in new pairs of images with an accuracy of about 80%, says Farid. Although the technique is currently specifically tuned to images of people, Farid says that the underlying algorithms could easily be adapted to analyse scientific images, using journal editors and scientists during the training process.
Farid notes that image manipulation is a growing problem in the scientific community, calling it "extremely disturbing”. He explains that it has become all too easy for some researchers to misrepresent their results, enhancing DNA bands in a gel, for example, or scrubbing out background blemishes, either to innocently make images look better or, in some cases, to skew the results deliberately.
Picture imperfect
It is not clear why scientific image fraud is a growing problem, says John Dahlberg, director of investigative oversight for the Office of Research Integrity in Rockville, Maryland, whose division investigates cases of alleged research misconduct. “It seems the scientific community is very aggressive about beautifying its images,” he says. “About 70% of our cases involve questioned images.”
See what we're tweeting about
16 Comments
Add CommentIf "unrealistic" body images cause eating disorders ALL cartoons should be immediately banned.
Reply | Report Abuse | Link to thisYou nailed it, Candide. "Unrealistic body images cause eating disorders" is the exact same thing as "The promotion of unrealistic body images in some ads and magazines is thought to have a role in triggering eating disorders." Spot on, truly.
Reply | Report Abuse | Link to this"It is not clear why scientific image fraud is a growing problem"
Reply | Report Abuse | Link to thisOh please ... 'funding competition' ??????
Digitally altered images should be equated to 'Lies'. Basically they are presenting the reader with something that is not true, but concocted.
I would suggest that there is a huge difference between exaggerated art expression and presenting 'accurate' information with favorably contorted visuals.
Reply | Report Abuse | Link to thisEating disorders affect 3% or less of Americans, yet 75% of Americans are overweight or obese. It seems the more pressing issue is the latter. I think images of overweight models should also carry a disclaimer, "Warning, being overweight or obese dramatically increases your risk of heart disease and diabetes." An overweight/obese girl modeling a dress is like a cowboy modeling with a cigarette. When did self-esteem, which should be based on aspects other than personal appearance, take precedence over personal health? We tell people that big is beautiful and worry about eating disorders and people thinking they're not skinny enough. Well, guess what - 75% of Americans aren't skinny enough!
Reply | Report Abuse | Link to this"...legitimate adjustments such as cropping and colour modification."
Reply | Report Abuse | Link to thisCropping can avoid showing certain parts of your body, such as a flabby stomach or blemishes. Color modification can be used to create fake tans (or bleach skin), change hair and eye color, and remove some blemishes by evening skin tones. Just because a graphic artist isn't increasing a model's bust size doesn't mean s/he is making "legitimate adjustments."
[Digitally altered images should be equated to 'Lies'. Basically they are presenting the reader with something that is not true, but concocted.]
Reply | Report Abuse | Link to thisSo correcting "red eye" for example is presenting something that is not true? Or is it rather correcting the difference between how the camera records it and the eye actually sees it? Personally I have never seen anyone in person with red pupils .. have you? While granted there are a plethora of fantasy images (whales floating through a room for example)there are many more images that are corrected for one simple reason, there is a fundamental difference between how a camera and the naked eye view an object .. or why else would it be necessary to have a host of lights one tiny little object whose details for example can be seen just fine in person?
you, sir, are spot on. and while people are happy to focus on the 3% of problems related to eating disorders (some of which i'm sure have nothing to do with distorted mag. images) i find it a much larger problem that a scientist would distort visual data such as a band of DNA on a gel. luckily the scientific process tends to weed out dishonest scientists
Reply | Report Abuse | Link to thisPerhaps the solution to scientific photo editing would be that peer review panels demand that the researchers include RAW images if they have done any kind of editing to their images. I have done a lot of scientific imaging and it can be helpful to clarify or emphasize aspects of an image, but I can see how it could be abused to create data where there is none, especially in something like a gel where subtle contrast distinctions are so key to accurate readings.
Reply | Report Abuse | Link to thisThe EXIF data is readily available in any graphic image and will typically contain information about the image's origin, such as whether it came from a camera or Photoshop. While the EXIF data can be directly altered, there is a LOT of data and some aspects cannot be altered from the original. That is, a true RAW image directly from a camera (without any alterations) can usually be verified as such. EXIF metadata contains as much as around 30 attributes. It would be tedious and very difficult for a researcher to perfectly alter the EXIF and other metadata of the file. Modern peer review increasingly relies on digital files, so maybe an editor or other official/staff at a journal could be assigned the task of verifying the EXIF data of required RAW images.
Perhaps if papers that are rejected due to image manipulation were not published, but the names of their authors, the titles of the papers, and the reasons that the papers were rejected (deliberate manipulation of images and data) were published, the negative publicity would serve as a deterrent.
Reply | Report Abuse | Link to thisThere needs to be some legislation on this, though I am less concerned about magazine covers than the many other more malevolent forms of image manipulation that take place. In science images have been basically stolen and re-used without proper citation. In highschools image manipulation is used to abuse students. And throughout the web images are altered to basically create lies. Criminal activities IMO.
Reply | Report Abuse | Link to thisI can only laugh at the Nanny State nonsense. I proudly proclaim that every image that I publish and print is always manipulated, modified, shape changed, layered, sharpened, de-sharpened, micro-managed, painted, spindled and mutilated..digitally or otherwise. It is called making art. When exactly did realism or the equivalent become a requirement for image making and surely for advertisements? It is to laugh. To blame an advertiser for a poor person with an eating disorder is questionable. I suggest that one look closer to home..to one's social contacts, parents, schools and personality disorders due to low self esteem and so forth. There have always been models of varying shapes and sizes throughout the past two centuries that depict idealised men and women but only in the past few decades is there an issue of self strarvation. I suggest it is not a simple case of blame the artist/designer and labelling such imagery is about as useful as labelling a ham as possibly carcinogenic. It is not going to stop a consumer from eating it nor from a consumer from looking at an advert. What it does do however is turn the world into a fascist information controlling superstate. Consider this. In Quebec, Canada, there is a sort of language police wherein it is illegal to post a sign on the front of a store that is not in French in much larger fontsizes. Say what? I am not exaggerating. This is the same sort of perverse use of thought police as would be the result of government clones making sure that a perfume add is pure and virginal and untouched and if it were, that it was labled as manipulated. Tell that to Picasso's ghost or that of any artist anywhere. Photography is not a data system but a language and thought police such as the Quebec language police is not needed nor welcomed by artists and designers. It is essentially the same as alcohol prohibition pushed by lunatics who will paint all ideas as dangerous...Spare us all.
Reply | Report Abuse | Link to thisI completely agree - we don't need disclaimers in magazines or any popular media, or it certainly shouldn't be written into our laws. If a privately owned publication wants to do it, that's their business.
Reply | Report Abuse | Link to thisMy only concern is with edited scientific images, but science is a self-correcting field that relies almost totally on peer review. There need not be any laws in this area, either.
As far as eating disorders and obesity, it would be better if we worked toward better education and giving people incentives to eat healthy. Most Americans now know what food is bad for you, and have significant access to nutritional information, often in the restaurant itself, but this has not changed anything. There are underlying causes for unhealthy diets, whether a person is eating too much or too little. Neither will be solved by regulating magazine ads.
It would be interesting to apply this to a certain president's "birth certificate."
Reply | Report Abuse | Link to thisWe need software to spot fraud in the comments of politicians. Those of us having a smartphone should be able to have a CNN/Fox News/MSNBC program running parallel with a TV program and the mobile phone app immediately shows the error in the statements of the politicians.
Reply | Report Abuse | Link to thisAre you trying to say that Spielberg's dinosaurs weren't real?
Reply | Report Abuse | Link to thisAfterall, they were members of the union. And, you have to admit that T-Rex moved like it was really there.
Personally, I believe.