Exam 2: Color in Image and Video

Early vacuum tube based color video cameras sensed color by placing Cyan and Yellow "stripes" in a diagonal pattern, leaving some places blank. In terms of R,G,B, what colors or color combinations where thus sensed by the camera?

Suppose we look at an image on the monitor. In the entire process of making that image, from acquisition to display, how many different functions of wavelength are involved in creating the sense impression we receive? Explain briefly.

Given chromaticity values $x, y$ , how do we reconstitute the tristimulus values $X, Y, Z$ ? Actually, this is an impossible question, since chromaticities remove the brightness from colors; therefore a proper question might be: given a target luminance value $Y$ (often taken to be 1), what is the relationship between $x, y, Y$ and $X, Y, Z$ ?

A blue object appears blue in sunlight since all blue light impinging on it is absorbed by it?

In a YIQ decomposition of the image parrots . gif for a parrot's face and beak, which are respectively whitish and blackish in the original colored image, we notice that in both the I and Q images these regions are almost the same dark gray level. Explain why that is the case.

A colorimeter has three controllers that can be independently adjusted, for color matching. Briefly explain how the color-matching functions might be measured using this device. (b) Is it possible to set the controllers to numbers (at least one of which is non-zero) that correspond to colors that cannot be seen by a human?

Suppose we decide to create a 5-ink printer, with colors CMYGK, where G is Chartreuse (Chartreuse is a real color - it doesn't matter what color it is). How many vertices appear in the printer color gamut? Why?

What color is the sun? Suppose we know the spectrum $E(\lambda)$ of sunlight. How would we calculate a screen display color, with 8 bits per color channel, for that spectrum?

Suppose the set of color-matching functions $\bar{x}(\lambda), \bar{y}(\lambda), \bar{z}(\lambda)$ , are given as a set of 31-vectors, representing values at $400 \mathrm{~nm}, 410 \mathrm{~nm}, 420 \mathrm{~nm}, \ldots, 700 \mathrm{~nm}$ . How would we visualize in 3-dimensional X, $\mathrm{Y}, \mathrm{Z}$ color space all possible colors that can be seen by humans?

What color light has the shortest wavelength: blue, green, yellow, or red?

In the simplest version of the median-cut algorithm, does it make any difference whether we assign bits in the order RGBRGBRG, or GBRGBRGB, etc. Explain. (b) Suppose we decide to quantize an 8-bit grayscale image down to just 2 bits of accuracy. What is the simplest way to do so? What ranges of byte values in the original image are mapped to what quantized values?