Patent focuses on deepening optical knowledge|
(Wednesday, March 19, 2003 - 01:04 EST)
Slashdot, the geek-friendly technology news website that is perhaps better-known by its punctuated abbreviation /. has picked up on a very interesting story from CDM Optics Inc.
As frequently happens when a company is "slashdotted", CDM's website has temporarily vanished off the face of the earth - but a wealth of background information on the technology they've been working on in cooperation with the University of Colorado can be found elsewhere. CDM was founded in 1996 by three individuals with close ties to the University - R.C. "Merc" Mercure (CEO), W. Thomas Cathey (President) and Edward Dowski (Vice President). The company aims to commercialise a technology which could have a dramatic impact on the photographic industry amongst others, called wavefront coding.
A patent application submitted by Dowski and Cogswell was accepted late last month (Titled "Wavefront coding phase contrast imaging systems", it is US patent #6,525,302) and assigned to The Regents of the University of Colorado. In turn, the University Technology Corporation has exclusively licensed the technology to CDM Optics Inc., which has been working further on bringing it to market.
The patent looks to solve a major problem experienced by current lens designs - that of depth-of-field. Any photographer will be familiar with the dilemma of whether to accept a narrow depth-of-field in a photograph and potentially find your subject out of focus, or close down your aperture to enlarge the depth-of-field at the expense of the brightness of the lens (either increasing noise or lengthening exposures and risking blurring from camera shake or subject movement).
Dowski and Cogswell noted that the problem with the current system is the method of controlling depth-of-field using a variable aperture. In effect, this solves the problem by reducing the amplitude or intensity of the light, bringing instead the other problems noted above. The new invention instead uses wavefront coding to change the phase of the light - providing an image which is effectively evenly unfocused over a certain distance range. This might seem like the last thing you'd want to do - deliberately capturing an unfocused image - but Dowski and Cogswell then take advantage of the exponential increases in computer power we've seen in the past, along with the uniformity of the lack of focus to recalculate the captured image and correct the focus.
The result is a much wider depth of field than can be achieved with traditional methods, without the attendant loss of light and potential exposure problems that brings. The technology is likely still quite some time away from having any effect on the digital cameras currently in development, but as it is further developed and refined for use in other industries - and there are a wide range of industries that could find potential advantages in a wider depth of field such as biometrics, microscopy, endoscopy, machine vision and more - it is likely to eventually bring advantages for traditional photography as well.
The potential advantages for photographers don't just stop at greater depth-of-field or greater lens brightness, though. Consider that with a greater depth of field, accuracy of focus might not be quite such an important factor - focusing systems could sacrifice some accuracy for increased focusing speed, which is yet another of the major issues photographers currently face. Increased depth of field could reduce the amount of movement required in a focusing system, or if depth of field could be increased enough, could even negate the need for power-hungry focusing mechanisms altogether, increasing camera battery life, reducing camera complexity and cost, increasing reliability, and so on.
Perhaps most exciting of all is the though that if a sufficient depth of field could be captured in the first place, there might be the potential for post-exposure focus compensation in much the same way as some RAW formats offer post-capture adjustment of white balance, exposure and so forth now. The thought of being able to click on an image on your computer and define which subjects need to be in focus would strongly appeal to many photographers. The curse of the perfect photo that was ruined because your camera elected to focus on the background instead of your loved one's face is something we could do without.
Whenever it recovers, CDM's website will likely yield more information on the new technology, and there are apparently a number of sample photos demonstrating the technology to be found there. We have also found a number of sample photos and simulated images on the University of Colorado website, along with articles from The Economist, Business 2, The International Society for Optical Engineering's oe Magazine, and The Denver Business Journal. Incidentally, we didn't notice it at the time but Carl Zeiss announced an agreement to use CDM's technology in its own products in September 2001. This was followed in August 2002 by Olympus announcing plans to use CDM's technology in its endoscopy products. We're sure other manufacturers are lining up to gain access to the technology as well, and CDM was apparently ranked the 8th fastest-growing company in Colorado late last year...
Further information can be found at the US Patent Office, and we'd recommend interested readers to look at US patent numbers 5,748,371 "Extended depth of field optical systems", 6,021,005 "Anti-aliasing apparatus and methods for optical imaging", 6,069,738 "Apparatus and methods for extending depth of field in image projection systems" and 6,525,302 "Wavefront coding phase contrast imaging systems". You might also want to search for the names of the inventors to find some other interesting patents, such as one entitled "Three-color imaging on each pixel for increased resolution" (#6,417,950) which looks rather interesting...