My Thoughts on Technology and Jamaica: University of Washington HyperCam - How Hyperspectral Imaging can improve Biometric Identification in Apple iPhones

Sunday, October 18, 2015

University of Washington HyperCam - How Hyperspectral Imaging can improve Biometric Identification in Apple iPhones

“When you look at a scene with a naked eye or a normal camera, you're mostly seeing colors. You can say, 'Oh, that's a pair of blue pants. With a hyperspectral camera, you're looking at the actual material that something is made of. You can see the difference between blue denim and blue cotton”

UW Computer Science and Engineering doctoral student lead author Mayank Goel commenting on the HyperCam

Our human eyes can only see in the visible spectrum, from Red to Voilet, which is a very small part of the Electromagnetic spectrum. But what if we could see everything? What would the world look like?

Researchers at University of Washington and Microsoft Research have developed a HyperCam that can see in every frequency of the electromagnetic spectrum as reported in the article “New camera captures invisible details, sees through the skin of fruit”, published October 15, 2015, by Michael Franco, CNET News.

The research was led by UW computer science and engineering doctoral student and lead author Mayank Goel, who led both the teams from the University of Washington and Microsoft Research as reported in the article in the article “Affordable camera reveals hidden details invisible to the naked eye”, published October 15, 2015 by Jennifer Langston, University of Washington.

His paper was presented at the UbiComp 2015 conference in Japan in which he outlined plans to make it possible for anyone to purchase with prices as low as US$800 and smartphone compatible version costing only US$50.

Hyperspectral imaging is not a new concept as NASA's Airborne Visible/Infrared Imaging Spectrometer (AVIRIS), 224 bands Imaging Spectrometer uses this technology. So it really does need to come down to Earth.

So aside from the possibility of yet another low cost camera component like the California Institute of Technology 3D Scanning camera component as retailed in my blog article entitled “@Caltech's portable 3D Scanning Device - How portable 3D Scanning using LIDAR gives a boost to 3D Printing revolution” , what else is so special about this Hyperspectral Imaging Camera?

How the HyperCam University of Washington and Microsoft Research works - Lens flares means special filters are needed

The HyperCam uses hyperspectral imaging, which captures all frequencies from across the Electromagnetic spectrum in the different frequency bands, almost like an X-Ray machine as noted in the article “HyperCam: $50 smartphone X-ray camera attachment can really get under your skin”, published October 16, 2015 By Mary-Ann Russon, International Business Times.

 Software then converts these bands to colours that we humans can see by comparing it against the original RGB (Red, Green Blue) image seen by the camera. The software algorithm then combines all these frequencies bands and removes unnecessary details and artifacts in the image.

The final combined image is used create an enhanced image, that reveals more texture about everyday objects than a regular RGB image.

An obvious application is seeing under women’s clothing. Though this may sound juvenile, obvious if you have a camera that can see in all frequency bands of the electromagnetic spectrum, (including X-rays?) it should be able to see through objects that would otherwise be optically opaque, like an X-Ray machine.

See where I’m going here TSA? Small than an X-Ray machine and built into Cameras, it could sport terrorist and identify bomb components before they had a chance to board and detonate them on a commercial airplane.

Alas, X-ray aren't naturally occurring in the environment, so it wouldn’t  be able to see through objects just reveal more surface detail for organic and inorganic objects. Also, since it’s basically able to see every frequency band in the spectrum, it can't handle bright light or lens flares very well.

The software can be tweaked to can remove lens flares, but when they occur they can damage the sensitive CCD (Charge Coupled Devices) as well as other specially designed sensor used in this type of camera. So special EM frequency filters are needed to protect the HyperCam’s components from damage.

How to see Yendi Phillipps naked - Improve spectrophotometry for Biometric identification in Apple iPhones

It would also reveal more surface detail and make it easy to identify materials using IR (Infrared) specrophotometry like the US$199 SCiO Portable IR Spectrometer developed by Consumer Physics as detailed in my blog article entitled “Consumer Physics US$199 SCiO Portable IR Spectrometer – Star Trek Tricorder that can scan the Molecular World”.

Based on that reasoning, it has more scientific and less juvenile uses, like spotting rotten fruit or fruit that is under-ripe in a Food process plant to quote Professor of Computer Science & Engineering and Electrical Engineering at University of Washington Shwetak Patel: “It's not there yet, but the way this hardware was built you can probably imagine putting it in a mobile phone. With this kind of camera, you could go to the grocery store and know what produce to pick by looking underneath the skin and seeing if there's anything wrong inside. It's like having a food safety app in your pocket.”

Like the SCiO Portable IR Spectrometer it can also identify the material compositing of materials based on the natural backscatter radiation coming from the material.

To quote Microsoft researcher Neel Joshi: “Existing systems are costly and hard to use, so we decided to create an inexpensive hyperspectral camera and explore these uses ourselves. After building the camera we just started pointing it at everyday objects -- really anything we could find in our homes and offices -- and we were amazed at all the hidden information it revealed”.

But most impressively, it might lead to a new form of biometric identification, as the HyperCam can used to create unique facial recognition patterns, being as it can see under your skin and possibly right through you as noted in “Microsoft, University Of Washington's HyperCam Could Give X-Ray Vision To Your Phone's Camera”, published October 16, 2015 by Anu Passary, Tech Times.

Thus with blood vessels included in the picture, unique picture of the person. Being as these veins are unique to hand, face or their entire body can be formed; faking your identity would become extremely hard indeed.

Something tells me this HyperCam will end upon in the next Apple iPhone. For now, I can only dream of using this to look at Yendi Phillips naked, as it should be able to see under bikini at least.....

No comments: