Answer to the Friday Whatsit for August 13, 2010
Much bafflement (is that a word?) from the Tiny Science reading community this week. No one managed to identify the tool represented in this image. April gave it a good try, throwing out all sorts of image processing ideas from the world of astronomy. But, since all her points are vanishingly small and very far apart, I can’t imagine she gets much use from a tool like this.
This is a distance map. According to the software documentation its a Euclidean distance map. There are other sorts of distance maps which are calculated differently, but all these maps do pretty much the same thing — help to measure distances between or within objects. This map would give you an idea of the maximum distance between object centers. The inner line (grayscale value = 0) runs down the geometric center of each object. The next row of pixels has grayscale value = 1, the next = 2 and so on until it runs into the expanding map from another object. The largest grayscale value in any part of the image is the maximum distance (in pixels) from center point to center point of the objects in the area. If you know what distance each pixel represents (at this magnification about 260 nanometers) you can know roughly how far apart two cells are. The smallest tick marks on a ruler are millimeters, there are a million nanometers in each millimeter.
I suppose you are interested in why I keep talking about grayscale when the image is clearly in color. Normally a distance map looks like the image on the right. Don’t see much of anything in the grayscale image? That’s OK, the human eye can’t distinguish shades of gray on the scale that a computer can. There are 72 shades of gray in that image. I had the computer put on a special color coding so that each gray level became a different color and the gradations were easier to see.