Science Podcast with Graduate Student John Hunt

In a podcast with Science Magazine reporter Kerry Klein asks John Hunt to describe the  innovative imaging tool for taking higher-efficiency microwave images described in   "Metamaterial Apertures for Computational Imaging" A full transcript of the podcast can be found below or by clicking here.

Play Sound

Host – Kerry Klein

Welcome to the Science Podcast for January 18th , 2013. I’m Kerry Klein [....]

Host – Kerry Klein

"In a digital camera, more megapixels means better resolutio—the ability to zoom in on an image and gain more information. But in many images, more pixels also means more redundant information and, thus, more required storage space. This week, John Hunt and colleagues describe a high-efficiency imaging tool that aims to collect only the bare essentials of an image. Where conventional cameras work pixel by pixel, the new tool instead breaks down a scene by wavelengths of light. The catch? The detector is made from so-called metamaterials--specialized substances which, so far, are only calibrated for the microwave part of the spectrum. I spoke with Hunt about the mechanics and practical uses of this technology."

Interviewee – John Hunt

"Microwave wavelengths are the wavelengths that your cell phone uses to communicate, for instance. We still think of them as a type of light. They’re not light that we can see, though. And this imager that we’ve designed in this experiment uses no moving parts, no lenses, and uses only a single detector. It’s equivalently a single pixel. This is made possible by combining two developing technologies. First, we use metamaterials that allow us unique control over light waves, and we use another technique called computational imaging, which generalizes how we think about and collect images."

Interviewer – Kerry Klein

"Okay, so let’s start with the basics. What exactly is a metamaterial and what sorts of metamaterials did you use here?"

Interviewee – John Hunt

"So you can think of a metamaterial as a type of composite, a composite material like, say, fiberglass. In the case of fiberglass, you combine two different materials – a woven glass thread cloth and a plastic resin – and by combining them and structuring them in a careful way, you come up with a new material that has different mechanical properties, different from and better than either of the two parts. So in metamaterials, we do the same sort of thing. We make a composite of different structures – different metals, different plastics –that gives us not mechanical properties but optical properties. We can control the way that light refracts and reflects through this material in unique ways. So one of the properties of metamaterials is that they tend to rapidly change how they respond to different colors or different frequencies of light. In previous experiments, this ha sometimes been a limitation. For instance, in the cloak – which is a pretty famous metamaterial device – it actually limits the operation range. It only works for one frequency, one color of light. But in the current metamaterial imager, we’re actually leveraging this behavior as a way to collect more information from the scene withouy using moving parts or without multiple pixels."

Interviewer – Kerry Klein

"So how do more conventional imaging technologies work, say, a hand-held point-andshoot camera?"

Interviewee – John Hunt

"In a basic point-and-shoot camera, you have a lens that focuses light from different parts of a scene to different pixels on the detector array, so that every point in the scene that you want to image is mapped by the lens to a different pixel. So if you want to have a million pixels in your final image, you have to have a million different detectors on your detector array. If you want to image at longer wavelengths and optical wavelengths, such as a microwave wavelength with this imaging system that we’ve designed, it’s made to work, you can’t use these millions and millions of pixels anymore because the resulting array of pixel detectors becomes far too large and costly to use easily. So, instead, what people typically do to image at microwave wavelengths is they take a single detector and they move it from point-to-point across sort of a virtual detector array, so that eventually they sample the light at every point that they would have put a pixel at if they could have made an array of many, many pixels. The problem with that approach is that it’s very slow to move this single detector, or array of detectors, across this plane. It requires complicated gears that are expensive and take up a lot of space."

Interviewer – Kerry Klein

"And how does your metamaterial aperture work in contrast to these more conventional technologies with their own sets of limitations?"

Interviewee – John Hunt

"The first difference is that there’s only a single detector, and we never move it. And the way that we make an image with this system is we have a metamaterial screen, which is covered in patches of metamaterial elements, which are each transparent to different wavelengths of light. So this means that for every color or wavelength of light coming from the scene, it’s sampled by different patches of the aperture before it gets to the detector. If you want to collect a lot of data from a scene, you have to, in a senser multiplex the way that you’re sensing. So one way of doing that, the way that it’s done in traditional cameras, is you have many different pixels. Pixels are spatially separated from each other. Instead of doing this sort of spatial detector multiplexing, what our system does is sort of frequency multiplexing so that each frequency or wavelength light that comes into that imaging system samples a different portion of the scene. And then we use some very elegant math, which is developed in the field of computation imaging, to turn that data into a two-dimensional picture of all the scattering elements in the scene."

Interviewer – Kerry Klein

"Taking a step back, you describe this metamaterial screen as being made up of patches that are receptive to different wavelengths of light."

Interviewee – John Hunt

"That’s right."

Interviewer – Kerry Klein

"How does this patchwork help process spatial information?"

Interviewee – John Hunt

"So the way, in general, that the metamaterial aperture encodes spatial information into our individual measurements is by focusing light from different points in the scene onto our single detector. And for every frequency that we tune our detector to, the metamaterial aperture focuses a different set of points from our scene down onto that detector. So we make a sequence of measurements for different frequencies, and we geta sequence of different intensity measurements that correspond to the sum of the points in the scene that are being focused onto the detector for each frequency."

Interviewer – Kerry Klein

"Ah. So you’re not measuring all wavelengths of light at the same time. Instead, you’re tuning the system to only detect one wavelength of light at a time, and then you’re running through these wavelengths sequentially to make a robust image."

Interviewee – John Hunt

"That’s exactly right. The detector is tuned to one frequency after the other sequentially, but because this is all electronic, it can be done very quickly. The metamaterials themselves are static; we don’t change them. But what we do change is the wavelengths of the light that we feed in to our metamaterial aperture."

Interviewer – Kerry Klein

"So what kinds of scenes have you been able to image so far?"

Interviewee – John Hunt

"So right now we’re imaging pretty controlled scenes. We’re imaging scenes that we’ve artificially constructed in rooms that have no reflection. So we cover the walls and floor and ceiling in a room with a non-reflective material. That way we can put things in the room and we can only see those objects; we don’t see any of the walls or other objects.This system works at microwave wavelengths, so you can’t see things like the sun, necessarily. What you can see are any metallic or shiny objects that reflect microwaves. In addition, right now the system only images in a single plane. It’s kind of like a radar plot. We have one dimension of range and one dimension of angle in our resulting images. We’re working right now on extending the system to make full threedimensional images where we would have two dimensions of angle and one dimension of range. And we’d be able to locate all scattering objects and shapes in a three-dimensional volume."

Interviewer – Kerry Klein

"So for right now, there’s no way to adjust the focus or depth of field of an image?"

Interviewee – John Hunt

"Well, in a sense we actually have a very large, perhaps infinite, depth of field. We can see objects at almost any range between one and five meters, for instance; we don’t have to focus at one depth. That’s one of the interesting advantages of this system."

Interviewer – Kerry Klein

"Does that mean then that all objects in the field of view are equally in focus, and that there’s no sense of depth at all?"

Interviewee – John Hunt

"Actually what we do have, one of our two dimensions in our image is a depth dimension. So we can tell where things are in distance, and we can tell where things are left and right. We can’t tell where things are up and down. And that’s because our aperture right now is a one-dimensional aperture. It’s just a thin strip. In order to have information about the up and down direction, we have to make that 1-D aperture a 2-D panel-type aperture, which we’re working on right now."

Interviewer – Kerry Klein

"And so how long does it take to actually collect the data and process an image?"

Interviewee – John Hunt

"The collection time is something like 50 milliseconds, and the processing time to generate an image from that collected data is approximately 50 milliseconds. So the time to capture and generate one frame is a hundred milliseconds, and we can do that at about 10 times a second."

Interviewer – Kerry Klein

"So what are the practical uses for this technology? How do you envision it ultimately being used?"

Interviewee – John Hunt

"So this kind of technology would be useful for any application where you’d like to have a cheap, small, microwave or infrared imaging system. So for instance, if you wanted to build an imager into the body of a car so you could do collision-avoidance imaging, or for security imaging at a checkpoint, if you wanted to just have an imager built onto a wall, or for instance if you wanted to have a cheap handheld device that could see through walls to find wires and pipes. Current systems cost millions of dollars to image at these frequencies, and this potentially could replace those systems with a very cheap,very lightweight, portable system."

Interviewer – Kerry Klein

"Do you see this kind of detector ever being viable at other wavelengths?"

Interviewee – John Hunt

"Well, some of the math and ideas would certainly apply, and already is being applied to other areas such as optical. But the current type of metamaterials, for instance, they don’t scale to optical frequencies. We could use these same ideas to make an optical imaging system, but we would have to change some of the hardware."

Interviewer – Kerry Klein

"Great. Okay, well John Hunt, thank you so much."

Interviewee – John Hunt

"Thank you."

Host – Kerry Klein

"John Hunt and colleagues write about innovative imaging in a Report this week."