Ratchet Up

Fauxtography

Cynthia Baron has a new book out, Adobe Photoshop Forensics. It's about systems being designed to detect Photoshopping. This from a Wired article:

A growing number of researchers and companies are looking for such signs of tampering in hopes of restoring credibility to photographs at a time when the name of a popular program for manipulating digital images has become a verb, Photoshopping. [...] Camera maker Canon Inc. sells a data-verification kit with some models. It can stamp digital photos with an invisible, mathematical summary of the image, such that even one tiny change will produce a mismatch and flag the photo as an alteration. These techniques are of interest to law-enforcement officials and defense attorneys because photographic evidence can make or break cases. News organizations also have been increasingly exploring ways to spot hoaxes. [...] The key [...] is to use tools in combination. A criminal or hoaxer might be sophisticated enough to defeat one technique, but not all at once. Fridrich's research takes advantage of the fact that all cameras have tiny flaws, so small they don't affect what the eye can see. For example, her software could analyze a set of photographs taken by the same camera and notice that a certain, defective pixel is always dark. Seeing that pixel light up would suggest an alteration. Dartmouth College professor Hany Farid, meanwhile, has developed a set of software tools he collectively calls Q-IF. He sells the programs for up to $25,000 a year. One tool looks for the use of clone stamp, a feature for duplicating or erasing objects in an image. Two cloned flowers would appear identical and lack expected blemishes. Another exploits how cameras capture color images. Color is a mixture of red, green and blue. Rather than have sensors that detect all three for each pixel, they generally alternate in a specific pattern. That pattern gets disrupted with airbrushing. Other techniques include looking for inconsistencies in lighting and shadows. [...] Photoshop already has a logging feature, which can track and record every change made along the way - standard procedure these days in law enforcement. [...] Adobe has no specific release schedule, though, on tamper-detection tools. The worry is that these same tools can help hoaxers test whether their changes escape notice."

[Source: Wired News]

Flugurator

Julius von Bismarck's Image Fulgurator is a device that senses the flash of other people's cameras going off and projects an image or message onto whatever they are taking a photo of, such as the Checkpoint Charlie sign in Berlin."

So reports Boing Boing. Here's a link to his site, which has been translated by Google—with some difficulty—from the German.

Image Search Engine: TinEye

A new service, still in private beta, that lets you search for a specific image on the web. No, not by category, as when you search for "donkey" in Google images, or Flickr. A specific image. Upload the image you are looking for to TinEye and it will supposedly locate all instances of that particular images on the web.

First use that comes to mind is looking for the author of an original image so that they can be properly credited. How often have you dragged an image into your pictures database then then found later that you would like to use it in a public context where it deserves to be credited. TinEye may be just the thing. If it works, that is. [Source: Robin Good]

Phillips' 3D TV

Phillips is now producing television sets the reproduce 3D images without the need for viewer glasses. Technology Review reports:

The technology uses image-processing software, plus display hardware that includes sheets of tiny lenses atop LCD screens. The lenses project slightly different images to viewers' left and right eyes, which the brain translates into a perception of depth. For now, the screens are expensive and not yet marketed for home use. But Philips, which first released the technology in 2006, is working on technical improvements that will make the screens better suited for the home. [...]

This isn't the first time that 3-D has made a splash. The early 1950s and early 1980s each saw their own fads. The 3D movies from the 1950s were filmed with two cameras, with the separate images then projected simultaneously. The familiar red-and-blue-lensed glasses were used to trick the eyes into interpreting color differences as distance. Modern 3-D movies employ more-sophisticated approaches, such as projecting the separate images in polarized light and using glasses with polarized lenses that filter out one image on each side.

But a combination of advances in computer image processing and industrial optics has allowed companies like Philips to develop their glasses-free technique.

As with earlier techniques, the illusion requires specially-created content to start with. In this case, a digital movie file effectively has two frames for each ordinary movie frame. The first is an ordinary color image, identical to what would be seen on a two-dimensional screen. A second frame, rather than showing a second offset view, encodes information about how viewers should perceive depth in the first frame. It appears as a grayscale version of the first, with white indicating foreground objects, black denoting deep background, and shades of gray indicating points in between

Then, special PC-based hardware and software--housed in the display itself--processes the pair of images as the video is played. The information in the second frame is used to transform the original color frame into nine separate images, each slightly offset from the last, as though the camera had been moved a few inches to the side each time. All nine are then sent to the screen.

To allow viewers to perceive these images, the LCD screens are overlaid with three-pixel-wide cylindrical lenses that direct the different images into side-by-side paths. A nearby viewer will see one of these images with each eye--the first and third, or third and fifth, for example--thus producing the illusion of image depth.

The multiple images allow viewers to walk around the viewing area--a cone about 20 degrees wide--without disturbing the 3-D illusion, says Philips product manager Erik van der Tol. This cone is duplicated several times on each screen, further widening the 3-D viewing area. [...]

As with any new technology, there are glitches. With the company's 42-inch screens, the 3-D effect works most effectively only up to a distance of about 12 feet, and if you view the screen at the boundary between the three "cones," you experience garbled images. In addition, the quality of ordinary two-dimensional images on the screens is diminished. Finally, a 42-inch screen will set you back $12,000 (prices on the new 52-inch and 22-inch models being released next week have not yet been specified)."

Look for this in the home in a couple of years.

Color / BW Optical Illusion

I have always loved visual illusions, but this is one I have never seen before. Amazing, to me at least.

Stare at the dot for 30 seconds. Then, without moving your eyes, move the mouse over the image. The image will look like it's in color until you move your eyes. "

[Via Daring Fireball]

Old Polaroid Down, New Polaroid Up

At about the same time that they are announcing the end of the production of Polaroid film and cameras, Polaroid is announcing its own new mini printer for your digital camera or phone. The printer fits in the palm of your hand and produces 2" x 3" borderless prints that are smudge-proof and water-resistant.

The Polaroid Digital Instant Mobile Photo Printer is the easiest way to print photos from a camera phone. Wirelessly transfer images from a Bluetooth-enabled camera phone to the Mobile Printer and print instantly. Using the built-in OBEX (object exchange profile) Bluetooth in the phone and the printer, connecting and printing becomes instant. Digital Camera Instantly printing images from a digital camera is as easy as connecting the USB cable to the PictBridge-enabled camera and printer. Once connected, follow the PictBridge screen on the camera display. Simply select what image to send to the printer and within 60 seconds, the print is ready to be shared."

[Source: Polaroid]

Elephant-Cam

Photographer John Downer attached a camera to the trunk of an elephant, producing some exceptional images.

Sky Ceilings

I'm not sure whether this is the new digital image or the old trompe-d'oeil. But I have a hankering to put a sky ceiling in my living room. Unfortunately I have an old school plaster-and-lath house, rather than the aluminum and fluorescent grid of new build construction. Still... [Via Boing Boing]

3D Video

Check out this immersive video. Once it is playing, mouse around to get an effect not unlike a Quicktime VR still image. It's made with eleven video streams arranged in a geodesic pattern, with a multi-lens camera like the one used for Google Maps' Street Views.

Camera Noise Signatures

If you take enough images with your digital camera, they can all be compared together and a unique signature can be determined. This means that even when you think that you are posting a photo anonymously to the internet, you are actually providing clues for the government to better tell who you are. The larger the sample size of images they have, the easier it is them to track down images coming from the same camera. Once they know all the images are coming from the same camera, all they then have to do is find that camera and take a picture to confirm it beyond a reasonable doubt.

It is important to remove this noise signature so that you cannot be tracked down. I cannot guarantee any of these methods will work beyond the shadow of a doubt because the woman doing research for the government on how to find the signature is very good. I can only promise that this will make their work more difficult." [Instructables]

21st Century Iwo Jimo

Ad for South African newspaper Die Burger [via Newsgrist]

Help Comparing Two Images

If you have ever wanted to know the differences between two images, imageDiff is the answer. There are a lot of tools out there for comparing text, whether it be source code files or office documents, but until now there hasn't been a good way to compare images.

Have you ever* looked at tw images and wondered if you are noticing all the differences in them?

* needed to know which file format is preserving your images the best?
* wanted to know how much a photograph has been changed when you compress it?
* needed to show someone the difference between two images, but found it difficult to explain in email?
* wondered whether someone else is using one of your images or just one that is similar?

ImageDiff can do all of this for you, and more. Whether you are a professional game designer, digital artist, photographer, web designer or just use a lot of digital artwork, imageDiff can help you be more confident that you know exactly what you are looking at.

Not only is imageDiff a great image comparison tool, it is available as both a standalone utility and integrated into our next generation source control and configuration management system – Evolution. If working with digital images is part of your profession, see how Evolution can save you time and make managing your digital assets easier.

Best of all, you can download imageDiff now for FREE! Or download an evaluation version of Evolution and find out for yourself how digital asset management with integrated image comparison can improve the way you work.
With imageDiff you can:

* Compare two images and see exactly what has changed, down to the pixel.
* Compare images in different formats with each other, including JPG, GIF, BMP, TIF, PNG and more.
* Compare images of different sizes, imageDiff will scale and compare them automatically.
* Choose between four different visualization filters to highlight differences the way you want.
* Overlay the original image to see exactly where changes have occurred.
* Report the percent of pixel and color change, a great way to see the effects of image compression.
* Adjust the comparison tolerance to ignore, or allow for, compression noise and artifacts.
* Switch between larger thumbnail views of the images to evaluate changes for yourself.
* Save the comparison image for reference or to be able to show others the image changes.
* Use the full-featured command line to run imageDiff from a script or plug it into other programs.
* Configure imageDiff as your default comparison program and set up a passthrough program so that non-image files are passed to a traditional text differ. [From IonForge]

Polaroid Printer Takes Beamed Images From Camphone

Millions of families once snapped Polaroid photographs and enjoyed passing around the newly minted prints on the spot, instead of waiting a week for them to be developed.

A tiny, battery-powered printer, top, from the company that was built on Edwin Land's Polaroid.

Now, Polaroid wants to conjure up those golden analog days of vast sales and instant gratification — this time with images captured by digital cameras and camera phones.

This fall, the company expects to market a hand-size printer that produces color snapshots in about 30 seconds.

Beam a photograph from a cellphone to the printer and, with a gentle purr, out comes the full-color print — completely formed and dry to the touch.

The printer, which connects wirelessly by Bluetooth to phones and by cable to cameras, will cost about $150. The images are 2 inches by 3 inches, the size of a credit card. The new printers are so lightweight that a Polaroid executive demonstrating them recently had three tucked unnoticeably into various pockets of his trim jacket, whipping them out as if he were Harpo Marx." [NY Times]

Googlehouse

Googlehouse is an online process that builds a house with images of domestic rooms (living room, tv room...) picked up on the internet using an image search engine.

Amped-Up Facial Recognition

You can take off that ninja mask now. A new facial-recognition algorithm created by researchers at the University of California at Berkeley and University of Illinois at Urbana-Champaign is able to recognize faces with 90-95 percent accuracy, even if the eyes, nose and mouth are obscured

.

"Most algorithms use what's known as meaningful facial features to recognize people -- things like the eyes, nose and mouth," says Allen Yang, a postdoctoral researcher at UC Berkeley's College of Engineering who developed the new algorithm. "But that's incredibly limiting because you're only looking at pixels from a designated portion of the face and those pixels end up being much smaller than the whole image. Our algorithm shows that you only need to randomly select pixels from anywhere on the face. If you select enough of them, you can produce extremely high accuracy." [...] Yang's algorithm ignores all but the most compelling match from one subject -- basically, its most confident choice. [...] The new technique could pave the way for completely new models for online advertising, new ways of annotating video and still images, and new techniques for monitoring and identifying people in public places.

Yang says he's already been approached by one startup (which he wouldn't name) interested in adopting this technique for what he calls "preannotation." For instance, this technology could automatically add family members' names to each image in a massive photo library, Yang says, saving you the trouble of flipping through thousands of photos to find that one of Uncle Bill.

It's also easy to imagine search engines like Google being interested in automatically recognizing the faces of the humans portrayed in publicly available photos, adding the image data to the textual information surrounding those photos to produce yet another dimension for targeting advertisements. Looking at a party photo of Johnny Depp on a fan site? Google could display advertisements for Sweeney Todd.

This new technique is also bound to raise a series of red flags for privacy advocates, since what Yang has developed is a highly accurate way of recognizing people even with occlusion or distortion.

With more and more cities, retailers and employers deploying security cameras in public places, it's only a matter of time before face-recognition technology like Yang's gets added to these cameras. Then the question will be not just who is watching you -- but whether they know exactly who you are." [From Wired]

Telling You No Makes It So Good

Strictly No Photography is a photo-sharing site for photographs taken where you are not allowed to take them. From the inside of the Kremlin to Kensington palace, from art galleries to war zones. Here you can see everything you've ever wanted to see that you're not supposed to. There are pictures that range from the ordinary to the profound. Whatever the content or the quality though we think that each one stands as a little piece of art in itself, as a little expression of personal liberty."

[Strictly No Photography via Fimoculous]

Face Detection Feature In Digital Cameras

If you have a new compact camera, take a peek at its specifications to see if it offers a face-detection setting. Typically, this option is in a camera’s autofocus (AF) menu. Face detection is particularly handy for candid shots, when you’re working quickly and are therefore more vulnerable to misfocused shots. It’s also a boon for flash photography. With face detection turned on, the flash doesn’t try to illuminate the whole room, just the people within range—cutting down on the nuclear blast effect.

Using Face Detection: With face detection turned on, the camera highlights faces on the LCD screen and then sets the focus and exposure for the subject.

Using the face-detection feature is fairly simple. As you compose your shot, your camera highlights the faces on the LCD screen and then gives you the green light to shoot. If the camera isn’t finding the person in your shot, the problem may be that it can’t see enough of his or her face. Face detection is much more effective when the camera can see both eyes of the subject; its accuracy diminishes greatly with profile shots. Also keep in mind that although face detection is fast, it isn’t instantaneous. For best results, compose your scene and then press the shutter button down halfway to activate face detection—this will give the camera time to adjust its settings appropriately. Once the camera shows that it has identified the subjects in your composition, press the shutter button down the rest of the way to make a perfect exposure.

Face detection is so simple that you may be tempted to leave it on all the time. But as with any setting, it’s not right for every situation. When photographing sporting events and landscapes, for example, you’ll probably get better results by switching to one of your camera’s other focusing settings. I recommend reserving face detection for family gatherings, weddings, and other people-oriented events." [From Macworld]

Mobile Phone Micoscopy

Researchers at the University of California, Berkeley, have developed a modular, high-magnification microscope attachment for cell phones. The device will enable health workers in remote, rural areas to take high-resolution images of a patient's blood cells using a cell-phone camera, and then transmit the photos to experts at medical centers.

The researchers hope that the innovation will help patients with blood disorders who live far from medical specialists get more accurately diagnosed and treated. "I wanted to make optical design relevant to today," says Daniel Fletcher, a professor of bioengineering at Berkeley. Fletcher's students found it relatively easy to integrate a simple arrangement of lenses with the cell-phone camera and transmit magnified images to a laptop using a Bluetooth attachment to the phone. The work prompted Fletcher to file a patent through the university and try to make a practical microscope. The researchers say that the cameras in late-model phones are capable of capturing all the details that a doctor would need to identify malaria parasites and cancer cells. [...]

The total cost of the first prototype, built from off-the-shelf components, was $75. The current version provides its own sample illumination from cheap, low-power LEDs. The device comes in two versions: with a magnification of about 5 times, for taking images of moles and rashes, and with a magnification of about 60 times, for capturing the details of blood cells and parasites. The higher-magnification model--the larger of the two--is roughly the size and shape of a roll of quarters. Both scopes attach to the phone with a modified belt clip.

"Microscopy is still considered the gold standard" for malaria diagnosis, says Katherine Herz, a medical doctor and a fellow in health policy at Stanford University. "If microscopy could be done with portable equipment ... [it] might be adopted far more widely and prove extremely useful." [From Technology Review]

Digg Pics

The recent launch of Digg's image section has given the images on Digg a home of their own. Building on our work in Digg Labs, Stamen Design built a structure for these images to flow in and out of, providing a view of the Digg community's visual browsing interests on a moment-to-moment basis.

Pics is divided into horizontal image strips, one for each section of Digg. As users vote on these images, they flow in from the left side of the screen. Over time, a mosaic of the images that people care most about builds up, and an overall impression of digg's users' activity becomes clear very quickly. Each of the images and sections can also be investigated on their own."

[More: Stamen]

Sees-Through-Clothing Cam

A camera that can 'see' explosives, drugs and weapons hidden under clothing from 25 metres has been invented. The ThruVision system could be deployed at airports, railway stations or other public spaces.

It is based on so-called 'terahertz', or T-ray, technology, normally used by astronomers to study dying stars. Although it is able to see through clothes it does not reveal 'body detail' or subject people to 'harmful radiation', according to the designers. [...]

Unlike current security systems that use X-rays, the ThruVision system exploits terahertz rays, or T-rays. This electromagnetic radiation is a form of low level energy emitted by all people and objects. These are able to pass through clothing, paper, ceramics and wood but are blocked by metal and water. The system works by collecting these waves and processing them to form an image which can reveal concealed objects. [...]

'You see a silhouette of the form but you don't see surface anatomical effects.'

In addition, the system does not involve any of the 'harmful radiation associated with traditional X-ray security screening', according to the firm."

[Source: BBC News]

Cameras As Pills

Technology that doctors expect will help detect precancerous cells faster and less painfully also could someday take cameras to parts of the body where no camera has gone before.

Cameras the size of pills could 'put eyes on tools' for laparoscopic surgery, snake inside a bile duct or fallopian tube, or weave their way deeper inside a person's lungs than any non-surgical device has been able to go.

Unlike a standard endoscope, which is almost a centimeter wide and can only be inserted into the esophagus after a patient is sedated, a new device invented at the University of Washington consists of seven fiber optic cables encased in a capsule about the size of a typical pain killer. [...]

The camera's 1.4-mm-thick tether allows the doctor to move the camera around and pull it back up once the five- or 10-minute test is finished. Human testing of the device is set to begin in about a month at the Seattle Veterans Administration hospital.

A larger, more expensive, but untethered pill camera was developed by an Israeli company in 2000 to test for intestinal cancer. Seibel said the disadvantages of the wireless camera are that doctors have no control over its path and it cannot be reused because it completes its voyage through the digestive system."

[Source: Wired]

High-Speed Photography to the Attosecond

Laser cameras are upping the ante on Harold Edgerton's bulett through the laying card of old:

"Capturing images of fleeting events—a horse's gallop, a bullet's impact, an electron's escape—is easy if you have the right equipment. Faster camera shutters used to be enough, but recently lasers have let physicists break the femto- and attosecond barriers, compressing the temporal resolution of images down to the time it takes light to cross a hydrogen atom."

[You'll find images at: Wired]

zZz is playing: Grip

A favorite self-reflexive online movie:

Video Camera Vest Transmits To Home Office

A UK company has launched a 'wearable CCTV' jacket which uses 3G cellular to transmit video images back to a central office. The company says that its WCCTV 3G Covert Backpack is ideal for agents who need evidential quality recording of suspects on the move. The backpack houses all of Wireless CCTV's body-worn technology in a covert package.

The central unit of the body-worn equipment uses a robust Compact Flash card for digital, evidential quality recording of agents' video and audio footage. Using 3G mobile phone technology, images and sound can be remotely monitored by a supervisor, who can evaluate the situation in real-time and despatch assistance or advise the agent on possible courses of action.

Wireless CCTV has also integrated an optional GPS receiver which tracks and maps the exact location of agents in the field. The precise position of multiple agents, combined with their respective live video and audio feeds, allows supervisors to assess a situation in real-time and make decisions accordingly.

A Panic Alarm button allows agents to notify the supervisor that they need immediate assistance. At the end of an operation, data can easily be backed up for evidential purposes, reducing the need for paperwork."

[Source: cellular-news]

Critter-Vision

I wrote a post quite a while ago about Sam Waterson when he visited Carleton for a talk. His critter-vision cameras are still in the news today with this piece from CNN:

Cameras as tiny as half an ounce are mounted on animals or plants. There's no specific length for the feature; it lasts until the camera falls off. The result is a unique perspective applauded by armchair naturalists in which the stars of the film are also the videographers.

'If people can see things from the animal and plant perspective, they are far less likely to harm them or their habitat, so that's how I present it,' Easterson said.

Earlier in his career as a video artist, Easterson put small cameras in strange places -- also with the goal of getting a different perspective. He put them in popcorn poppers and washers and dryers to show what those domestic appliances looked like from the inside out. Seeing through sheep's eyes

But equipping a small flock of sheep with cameras in 1998 changed everything for Easterson, he said. The Walker Art Center in Minneapolis, Minnesota, commissioned him to tape some sheep as they 'mowed the lawn' in a park. Easterson said he learned a lot about his new craft and the nature of animals.

'The first thing I realized was how intelligent and aware they are,' he said. 'The project changed everything in terms of respect.

'I was shocked to realize all the other animals in the flock could tell that this one sheep with the camera had been 'altered' in some way. She kept trying to enter, and they kept treating her as an outcast. I also learned sheep can run very fast and fences are not as sturdy as you think.'

That project was called 'A Sheep in Wolf's Clothing.'

Easterson takes small cameras designed for law enforcement and surveillance and modifies them for the animals or plants he's aiming to study. He changes them mostly by taking most of the weight off. For most smaller animals, the video usually is beamed wirelessly for recording.

Most of his work now comes from museums, nature centers and art centers eager to give the public an animal's-eye view. The artist said he's turned down requests from some broadcasters anxious to get gritty or confrontational pictures for 'reality'-type shows.

While not a scientist, Easterson works with researchers by offering details that they may have never seen.

'There really is data to see in these pictures,' he said. 'Take the armadillo. You can listen to its breathing patterns. You can watch closely the rotation of its ears as it encounters new things.'

While networks such as Animal Planet may focus on exotic international shoots, Easterson tries to include an animal native to the area where he's presenting an exhibit. In Florida, he equipped an alligator with the cameras, and in the Southwest, he chose a desert tarantula. A tumble weed's perspective

While it's easy to see the intrigue of animal behavior from an animal's perspective, it could be a stretch for some to consider a plant's view.

'It's maybe a funny idea to think that plants have perception, but I think they do,' Easterson said.

He said his most challenging plant shoot was driving along the desert in a rental car trying to keep up with a tumbling tumbleweed.

Easterson said no animals have been hurt working as videographers. And he's worked with bioethics groups and telemetry experts who are experienced in tracking animals with radio collars.

'I really would like to be on the radar screen, to improve my techniques,' he said.

In the long term, Easterson said he hopes to create a library of hundreds of animals, from the unique 'first animal' perspective." (Via CNN.)