January 6, 2011 Picture of the Week: Explaining 'the iPhone Effect'

Email this article |Print this article

copyright © Jerry Huether Used with permission

Last week, Jerry Huether submitted the photo at right (shot over Oregon looking toward Mt. Shasta), along with this comment:

What I really want is to know what caused the illusion of the prop blade detaching from hub. The background would be blurred if it was such a long exposure — and descent/climb was involved. Any ideas??

... and we thought that would be a good excuse to put the question to AVweb readers, which we did with a discreet link in last week's "POTW" slideshow.

We've often seen the effect here at AVweb world headquarters — most notably on photos shot with the built-in camera on the iPhone. Initially, we dubbed it "the iPhone effect," since frozen parallel prop blades didn't seem to occur on photos from (ahem) "proper" digital or film cameras, and we could never get it to happen on our pre-iPhone camera phones. But as the months have rolled by, we've seen the effect show up on other camphones, particularly those rascally smartphones, which (we assume) process digital images a little differently than the simple cameras plugged into old school phone that don't have a sophisticated operating system. That's all we knew about the iPhone effect — until our trusty readers deluged us with info and links.

The effect is caused by the time taken to "scan" the picture, as opposed to an "instant" picture taken by most digital cameras. The prop moves a bit while each scan takes place.

Bruce Marshall

Digital cameras don't capture frames as a film camera does using a shutter; it is a scanned line composite — like a TV displays a picture. The blades of the propeller are moving so fast the the blade position has changed between scan lines and creates that line effect.

Tim Wolf

This is a common phenomenon in cheaper iPhone cameras that have a rolling shutter. It is a type of aliasing. Aliasing occurs when the sample frequency of the changing image is insufficient to capture the image before it changes. Obviously, a propeller is moving very fast, and requires a very fast shutter speed to capture it when it is moving. Since the iPhone camera takes multiple exposures in a single photo, it tends to capture the propeller blade at multiple locations. That results in photos like this one.

A similar phenomenon has been seen in motion pictures for decades. Have you ever noticed a wheel on an automobile filmed for a movie? Have you ever seen those wheels turning backwards when the automobile is moving froward? This is another example of aliasing.

Jim Perkins

I often field this question in video discussions on BackcountryPilot.org, because the horizontal blind effect is common in this new crop of 30- and 60-frames-per-second HD video cameras, which are becoming increasingly popular to use for shooting cockpit video through the propellers of single-engine aircraft.

The effect is caused by the method though which each frame of the video is rendered by the image sensor, using what is called "progressive scan," where each line of the video is written in rapid succession until a single image is formed. Imagine your inkjet printer, making a single image line by line, from left to right, top to bottom. The prop moves too fast for your slow printer to capture its position accurately before it has reindexed its position in relation to the image rendering. High frame rate video reveals this scanning artifact, and what we know is a prop blade radiating out from the spinner is morphed into the illusion of magic horizontal floating bars.

Zane Jacobson

Digital cameras record images in many different ways. Some use a separate mechanical shutter to allow light to affect the individual photo-sensors for a predetermined period of time. Others turn the sensors on electronically and simultaneously for a predetermined period of time. Yet others extract information from the sensor by sequentially transferring data starting at the top of the frame until they get to the bottom.

When it comes to digital cameras built into phones, the situation is more variable. These cameras are most often equipped with CMOS (as opposed to CCD) sensor arrays and these lend themselves to line-by-line extraction or progressive image data. The camera that took the above photograph was equipped with just such a method for image extraction. Sometimes this is called a "rolling" shutter approach.

Ary Glantz

[Another] example is when you take photos of something in your house, and the TV is on in the background. You will often see a line across the TV, where the image is garbled, or even blank, in the final photo. [This is because] the electron beam that updates the image on the TV screen refreshes about 60 times per second. So a photo shot at 1/60th of a second captures a clean image, but a photo taken at a slower speed will capture part of the previous image, the "updating beam," and the new image — hence, garbled images or blank spots show up in the photo.

Unlike the TV example, there is no "magic speed" for clearly capturing a prop in a digital photo. It depends on the speed of the prop's rotation, the curve of the prop's blade, the shutter speed of the camera, and the pattern used when the digital camera's light sensor records the image. IPhones use a different pattern from Verizon phones. Nikons use a different pattern than Kodak or Minolta. And still photos capture light in a different pattern than digital video cameras. So you can play with your camera's settings (and the prop's speed) to minimize — or maximize! — this effect.

Remember how amazed you were the first time you saw a high-speed photo of a bullet breaking a ballon or an apple? It was the first time you "saw" what really happens — your sense of what is real being altered. This image aliasing effect is the opposite: The image is real, but you "know" a propeller can't do that. It makes these photos very interesting to look at!

Greg Wein

Pretty cool, eh?

We received about thirty e-mails detailing the effect in various degrees of detail, but these were some of our favorite comments and explanations. And for those of you still having a little trouble wrapping your brain about the several sets of variables that can affect "the iPhone effect," here are a handful of links that will lay it out for you, complete with videos, photos, and diagrams:

• Andew Davidhazy's excellent explanation and examples (via Blake)
• A quick explanation from the Scalar Motion blog, complete with an incredible visual explanation video (via Bill)
• More videos and an enjoyable blog on the methodology of sussing it out from Duke student Alex Beutel (via Greg Wein)
• Gizmodo's great example pic (and explanation) (via Waner Del Rosario)

Enjoy! And thanks to everyone who took a moment to chime in.