Midterm Exam 1 Review Spring 2003: Picture concepts


	Welcome to CS1315. Click on the python to add comments.

This page removed for FERPA compliance

Midterm Exam 1 Review Spring 2003: Picture concepts

Try the problem here from Sp 2003 Midterm Review #1

You can't see the individual red, green, and blue dots for each picture because they are too small to see. Our eyes blend them together to make them one color. You have to use some high-powered magnification devices (i.e. microscope) to see each of these individual dots. The numbers we use to encode the colors are the intensity of each color. Each pixel is made up of these triplet values that allow us to encode the intensity of each color component that we want.
Luminance is the way our eyes perceive colors. We don't see colors as they actually are, but a luminance perception of them. Our eyes are REMOVED sensitive to some colors than others. They percieve weighted values of each color. Those weights are in the book as .299 for Red, .587 for Green, and .114 for Blue. If we use those weights to change the encoding values for RGB in a picture, we can get a greyscale that's truer to our eyes.
Student116

Luminance isn't about colors. In fact, it's about the opposite. Mark Guzdial

Luminance is our perception of the darkness or lightness of the color. Much of our luminance perception is based on comparison to backgrounds, not raw values. This is how we see "borders" between black and gray for example.

I'm sorry, I forgot: black and white aren't colors- they're shades. But yes, I understand what Katie is saying, luminosity deals with the lightness and darkness of a color. This, once again, is the shade (shades, as I remember from art, are different versions of the color- just lighter or darker). I know, in the example in the book, you talk about our eyes percieving blue as darker than red. This can only be true, however, if the blue is a relatively dark blue (not baby blue). Red could be percieved darker than blue, however, if (a) the blue is a light blue, or (b) the red is a dark red, such as burgandy.
Student116

You're right, Rebecca, but in general, we perceive blue as darker than red. Even a dark blue can give off REMOVED light than a light red, but we perceive the dark blue as "darker." Our perception of luminosity is not directly linked to the amount of light being reflected off a surface. (We didn't get into it, but again, it's a matter of ratios. We note luminosity as a contrast with its surroundings.) Mark Guzdial

I find it interesting that (0,0,0) is black (no color) and (255,255,255) is white; when in reality, black is the absorption (or presence) of all colors and white is the reflection (or lack their of) of all colors. That's why black is always "hot" to wear, and white is often "blinding" to wear.
Student116

Good observation! That's because the values correspond to amount of light, not paint. (0,0,0) is no light => black. (255,255,255) is a mix of maximum red, green, and blue – all the colors mixed together => white. Mark Guzdial

So, white would be the presence of all light because each color reflects light back. In other words, white contains every color because all colors are reflected back making it a "colder" color to wear. Also, black must be a lack of light because no light(color) is reflected back. This makes black "hotter" because none of the light is reflected back, instead it's absorbed. Got it!
Student116

Question, if luminosity is a contrast with its surroundings, would luminosity be one of the ways in which we distinguish distance? Meaning, does it help us to distinguish distance?
Student116

I think so. The book I was reading on the psychophysics of sight says that we basically have two systems in our brains: the "WHERE" system (where luminosity plays a significant role) and the "WHAT" system (where color plays the significant role). Distance is part of "WHERE." We also use stereoscopy (the fact that we have two eyes), perspective, and other clues for distance. Mark Guzdial

I'm really having trouble understanding number 3. Does anyone have a clue or have some idea how to get me thinkin in the right direction?
REMOVEDsan Holloman

Okay, so there are 1024x768 (=786,432) pixels in the monitor image and 320x240 (=76,800) pixels in the cheap picture. If we take the difference, we get 709,632 pixels. There are 8bits per color component or 24 bits per pixel. 709,632pixelsRebecca REMOVED_
It's funny, you were posting the same time I was! I think that that's right._
That makes sense to me. thanks
REMOVEDsan

You're welcome!_
p.s.
Hey, but don't blame me if it's wrong, it just makes sense!Rebecca REMOVED

In class, he demonstrated this problem differently. Just letting you know.

well how did he demonstrate it?

Things can be demonstrated differently, but do they come to the same conclusion? Like in Algebra, there's tons of ways of stating the same thing. So, how did he do it in class?
Student116

In class did he go (1024768) x 3 = 2359296
- (320240) x 3 = 230400
————————-
2128896 ????
Someone who attended monday's class showed this to me tonight–hopefuly it is right

Yes, that's what he did in class. The three corresponds to the number of bytes. When Rebecca did it, she used 24, which is the number of bits. (8 bits per byte) When Rebecca divided to go from bits to bytes, she only divided by three, which is the number of bytes present per pixel, not the number of bits per byte.

So, you're saying:
(1024x768)pixels squared x 3bytes=2,359,296 bytes
Yup! I started with the smallest unit; however, I messed up converting to the larger one. So, my bits were right, just not my bytes.
Student116

Link to this Page

Sp 2003 Midterm Review #1 last edited on 5 August 2003 at 2:36 pm by user-1120spk.dsl.mindspring.com