There is an almost infinite number of colors we could use for the false coloring by spinning the color wheel, but it’s easiest to pick a primary color (red, green, or blue) and use its complementary color for the opposite. Some colors are harder to distinguish against bright backgrounds; think about picking out a bright yellow object against a bright white background. While other colors are harder to pick out against dark backgrounds; blue on black, for example. Is there an optimal choice that’s also inexpensive to compute?

Let’s find out. Here’s what happens when you perform the “What’s different?” test using the three obvious choices of red/cyan, green/magenta, and blue/yellow:

(Click any image to see it larger and/or full-size.)

The blue/yellow version is not very good because it suffers from the problems listed above, while the green/magenta and red/cyan false colorings seem to produce fairly equivalent results. What to do? The ever-practical Loren suggested that we see what happens if we ask people with color-blindness color confusion. Turns out, we know a couple of such folks, but they were in meetings. What to do?

Oh, that’s right!! A while back, I wrote some MATLAB code that simulates the two most common forms of anomalous color vision. Let’s run that and see what happens. (The speckling in the images is an artifact of the conversion and doesn’t represent how they actually look to people with protanopia or deuteranopia.)

The results are a no-brainer! (Sorry about the pun.) While the red/cyan and green/magenta images are very similar for the majority of us with normal color vision, the red/cyan images become very difficult to use for people with any kind of anomalous color vision. Even though the green/magenta starts to look a lot like the blue/yellow row, which has its own issues of color discrimination, it makes a very good compromise. Green/magenta it is.

I’m in California attending the IS&T/SPIE Electronic Imaging symposium. Today I took a short-course about “Perception, Cognition, and Next Generation Imaging.” I feel fulfilled by the first two parts; but the last one was a bit fuzzy. This evening, while watching Netflix, I figured out what conference presentations I plan to take in over the next few days. GPUs and parallel computing, high dynamic-range imaging, image quality, and maybe even a session called “The Dark Side of Color.”

But that all starts tomorrow.

I’m only gonna say this once while I’m here in California, because if I were in New England right now I would hate to be reminded even more that it’s winter. The weather is unbelievably beautiful here in San Francisco. When I did my weekly long run this afternoon, I felt something that I hadn’t in a long time: the 60s.

Today’s run was wonderful. The front desk staff at the hotel gave me a pocket-sized card with a couple of running routes that I could stitch together to get very close to the nine miles that were on today’s docket. But I got a little confused leaving the hotel, so I decided to keep following the trail along the Bay until I got to the far end of Coyote Point Park or 45 minutes in, whichever came first; then I would turn around.

I was feeling fast and figured that running 85 minutes would definitely be enough to have put the distance in. So imagine my surprise when I mapped the run later to learn that I had gone 10 miles, averaging a very brisk (for me) 8:30/mile. It’s amazing what happens when there aren’t multiple mile-long hills on the course. Today, for the first time, I’m starting to feel more confident that I might actually be able to attain the somewhat arbitrary stretch goal that I set for myself when I decided six weeks ago to train for the New Bedford Half Marathon.

(By the way, even though CGM still isn’t really very accurate when I’m exercising, I learned some things during my training last Friday that have improved the accuracy greatly the rest of the time. I like it. Here’s hoping it helps me get my A1c and my confidence where I want them soon. Oh, and one more thing; I signed up to do a sprint triathlon in early May in Hopkinton. G-d help me. More about both of those topics and about yesterday’s gallery hopping excursion later.)

And I miss Lisa terribly.

• I can’t decide whether Arcade Fire’s new album, “Suburbs,” is completely, utterly pretentious and lacking in fun, or if that’s me I’m thinking about.
• The second week of August may be the second best commuting week of the year. It has felt like the week between Christmas and New Years.
• The reception areas of Newton-Wellesley Hospital (NWH) are under construction, and the architects created a display of the materials they’re using. I like that a lot.
• Phlebotomists, who specialize in doing something inherently painful with a minimum amount of discomfort, aren’t paid well enough. I’ve been poked many times, and the ones who do it well really are amazing.
• The NWH lab dedicated to drawing blood is extremely quick. It’s where I prefer to go. It opens at 8:30.
• At 7:00 the main hospital lab claimed a 30 minute wait, but it was really an hour-long wait for 60 seconds of actual medical procedures.
• Some days I’m really eager to get to work and finish up what I was working on the day before. Today was one of those days.
• In early April, Sports Illustrated predicted the Chicago Cubs would finish second in the NL Central, with a record of 81-81. To make that happen, the Cubs will have to go 33-15 for the rest of the season. The Cubs also have an estimated payroll of $137M for the season, which is $100M more than the team one behind them, the Pittsburgh Pirates. (The Pirates!)
• I should have brought a book with me to the lab. I just finished reading about platypuses and have started reading about Romantic science.
• I was smarter during the company meeting. Now I know a lot more about “Black-point compensation: theory and application” and ICC color profile rendering intents than I did yesterday.

And now it’s time to muck around with run-length encoding.

Image Engineering: Dietmar Wueller has a great company that makes test equipment, targets, and software. I’ve had the good fortune of working a booth in a trade show next to them (more than once). I trust their expertise completely.

WyoFOTO, LLC: My mother and step-father are photographers and have a little web site of their own.

Trek 2.1 Road Bike: Every month brings a new event to distract me from posting here. In June, it’s my new bike.

Twitter: With millions of users, Twitter doesn’t really need me to get the message out. I think it’s a nice way to follow a small number of people who post links to photography, design, and software engineering articles/blogs/etc. that might interest me. Others use it differently — sometimes I do, too.

Flickr: They don’t need my advertising either.

The Wire: The final three episodes of “The Wire” should arrive in my mailbox today (via Netflix). If you’ve ever been hooked on the show, you’ll know another reason why I’m not posting more.

As usual, I learned something new and talked to a lot of people about imaging and MATLAB. Last year, the new-to-me things concerned image quality for mobile devices and image forensics. This year’s big surprises were 3-D imaging, which apparently is going mainstream. (According to Andrew Woods, the chair of the Stereoscopic Displays and Applications conference, there are already two million televisions in people’s homes which are capable of displaying 3D video.) Every year that I attend, I seem to get a different view of the conference; so I’m cautious to read too much into any given year. But somehow, 3D felt different this time.

The exhibitors behind us were demonstrating their more advanced 3D displays and video capture systems, and the person across from us was selling 3D notecards and prints. So my coworker and I thought we’d get in on the action during the waning hours of the exhibit. We decided to try our hand at making an anaglyph, one of the easier kinds of 3D images to make (along with stereograms, which are a bit lo-fi for MATLAB, after all).

The recipe calls for combining two images of the same scene from slightly different perspectives using a simple function that combines the red channel from the left image and the green and blue channels from the right image. And of course, you need the funny glasses. First, the images that we made with my camera phone:

Left image

Right image

Using some MATLAB code that we found on MATLAB Central, you get the following image:

The anaglyph

(Unfortunately, I can’t help you with the funny glasses. Go get yer own.)

I am told the results are “okay for a cha-cha picture.” I don’t have binocular vision, so what do I know?

Have fun. (And don’t be like the guy who said that the 1.57% of the population with monocular vision aren’t really worth worrying about.)

(Thanks to Steve on Image Processing for the link.)

Before I give you some scanty details about Hooman Mohseni‘s work into single-photon detectors, I’ll do what he did in his presentation and discuss rod cells, the eye’s own single-photon detectors.

The rod cells in your eye are almost perfect single-photon detectors. When a photon of light in the visible range strikes a rod cell, it causes a chemical reaction with rhodopsin, a photo pigment. The rhodopsin molecule changes shape, closing an aperture in the rod cell wall. This partially interrupts the flow of charged sodium ions into the rod cell, and an electrical charge builds up. The charge causes an impulse that is transmitted to the rest of the visual system, possibly causing the sensation of seeing light. There are several layers of rhodopsin in the cell, allowing a greater impulse when more photons strike it. (You can read more details about rhodopsin and its paramour retinal if you want.)

Each rod cell can detect a single photon event, and rods are photon number resolving, meaning that the visual system can use the impulse to distinguish whether one photon, two photons or twenty photons struck the rod cell. And the false count rate is very low, approximately one per 100 seconds. And all of this happens at low electricity levels. Pretty cool, huh?

Compare that to imaging sensors like CCDs, where at very low illumination levels noise overwhelms the actual signal. Even very accurate detectors that can detect single photons reliably (such as photo-multiplier tubes) require large amounts of voltage and usually can’t resolve the difference between one or two photons. Signal-to-noise ratios have been pretty bad so far, too.

Prof. Mohseni’s group at Northwestern has been using nanotechnology fabrication techniques to create a “focalized carrier augmented sensor (FOCUS)” which can transform a single photon event into an output of about 1,000 electrons, at very low voltages. Essentially they have produced something like a rod cell.

Their detector works really well, but nanofabrication is hard and slow. They can’t even image the small, tube-like sensors because they are too delicate; for example, atomic force microscopy — a very gentle technique — destroys the tubes. Nevertheless, this may be the future of things like night vision, positron-emission tomography, and so on.

Yasuyo Ichihara of Kogakuin University presented some research that she and her colleagues performed on what many of us call “color blindness” or “color deficiency.” She would prefer that we say these differently sighted folks have “color confusion,” which I’ll probably try since it’s slightly more evocative than “color deficiency.” And from a PC point of view, it may not be nice to stagmatize roughly 5% of the population as deficient, etc.

Anyway, they presented research of how their lab applied color universal design to remake the map and timetables of the Tokyo subway system to be friendlier for people with both protanopia and deuteranopia. It turns out that the color confusion can be quite simply modeled by placing two iso-chromaticy lines on the standard x-y chromaticity diagram. Along those lines, all colors appear to be the same hue; so it’s not too hard to pick colors that are maximally contrasting by avoiding the iso-lines. They picked four colors, which they claim are suitable for many print applications: black, orange-red, bluish-green, and a more pure blue.

Color universal design doesn’t stop at picking the right colors. Adding non-color information — such as letters, underlining, framing, or other symbols — proved to be very useful aids to comprehension, too.

All of this reminded me that awhile back I implemented a color deficiency simulator in MATLAB based on work by Hans Brettel. I’m making these functions available here in the hopes that they’re useful and that you actually use them when designing something colorful. Simply download the protanopia and deuteranopia simulator M-files and the associated data file. (You will need the Image Processing Toolbox, since these functions use rgb2ind.)

Let’s see what the peppers image I created some years ago looks like to people with color confusion:

rgb = imread('peppers.png');
figure; imshow(rgb); title('Normal color vision')
figure; imshow(deuteranopia(rgb)); title('Deuteranopia')
figure; imshow(protanopia(rgb)); title('Protanopia')

I find this topic really interesting for a few reasons. (1) I’m a photographer who believes that all images contain a seed of untruth but feel ambivalent about what that means in our information age where images form the basis of what most people “know.” (2) Practical applications of image processing are always interesting. (3) It’s fresh in my mind, since I recently helped someone show that an Ethiopian passport was a fraud. (Image processing played only a small part — the content and font on the machine readable section didn’t match the international standard — but showing that parts were pasted in digitally helped create a preponderance of evidence.)

Hany Farid of Dartmouth gave the first plenary session: “Digital Forensics”. Prof. Farid started by quoting a science journal editor’s staggering statistic: 20-30% of submitted journal entries need images resubmitted because of “inappropriate image manipulation.” His lab’s work aims to out these digital forgeries. Here are some techniques that you can use to identify likely forgeries.

JPEG quality tables (Q-Tables) — These 8-by-8 tables of quantization values are stored in each JPEG file and used to decompress the images. Something I did not know was that most camera vendors use unique Q-tables and that they frequently change them when they introduce new camera models. Photoshop, on the other hand, has not changed their Q-tables since version 1. So you can extract these values from a file and see if it has been saved by Photoshop, which might hint at manipulation.

Cloning — Partition the image into blocks, do principal component analysis, lexigraphically sort the results, do region growing, and look for similar regions.

Resampling (shrinking, growing, rotating, etc.) — Use statistical correlation to look for simple interpolation between values.

Included objects frequently have a different color filter array (CFA) pattern than the rest of the image, possibly because of resizing or different in-camera decoding. Also you can create a vector field of the chromatic aberration color fringing throughout the image; inserted parts will likely have vectors pointing the wrong direction.

In addition, the human visual system (HVS) doesn’t easily notice subtle differences in lighting or shadows in parts of a composite image, but Farid laid out some techniques for estimating lighting direction. One impressive method involves determining the direction of the light (in 3-space, mind you) by looking at the specular highlights in the eye. (See Micah Kimo Johnson’s thesis for full details.)

During the Q&A, one wag asked the question on many of our minds. Are there tools or techniques available to mask all of these forgery detection techniques? Apparently yes, since Prof. Farid consults with Adobe on issues of making more realistic photo manipulations.

Update — 10 August 2008: You can see a thirteen minute video with Dr. Farid at PBS’s Nova web site. (Thanks to Jeff Tranberry.)