a)   A 24-bit RGB image has size 200 x 200 pixels. The image is stored 
     in memory as four integer arrays - one array for the alpha channel 
     value, and one array for each of the r, g, b values. 

     You want to reduce the memory requirements of the image. To do this, 
     you convert the image to grayscale. You find that the memory remains 
     the same. 

     Explain why this is so.

     (4 marks)

b)   Explain how saving the image in part (a) in packed pixel format will 
     save memory. 

     Draw a diagram to illustrate your answer, including the alpha channel 
     value.

     (4 marks)

c)   How much memory will be saved for the image in part (a) if it is saved 
     in packed pixel format, as in part (b).

     Show your working.

     (2 marks)

==============

A)  The source code for a threshold operation requires some form of 
    conditional statement, eg. an if statement. 

    Explain what this conditional statement does. Do NOT write code 
    – explain your answer in words.

    (3 marks) 

B)  Give ONE example of a context in which thresholding an image is 
    useful.

    An example would be handwriting recognition. (Do NOT use this 
    example in your answer!)

    Explain your answer.

    (4 marks) 

=============

A)Sound is streamed across the network with the following characteristics: 
  
  sampling rate of 44100 Hz
  4 bytes per sample
  mono

  How many bytes are streamed per second?

  Show your working.

(5 marks)

B)The noise level of Noise A is 128dB. The noise level of Noise B is 80dB. 

  How many times greater is the noise level of Noise A compared to the 
  noise level of Noise B?

  Show your working.

(5 marks)
============

C1. A grayscale image has the following range of pixel values. 

Value is the pixel intensities from 0 (min) – 9 (max); 
N is the number of pixels with a given value.

Value	N

0	35	
1	26
2	15
3	15
4	10
5	4
6	15
7	14
8	21
9	25	
	
Equalise this image. You must show the following in a table:

• Current cumulative frequencies of the pixel values
• Mapped pixel values
• Rounded pixel values
• New N for each pixel value
• New cumulative frequencies of the pixel values

You must also give the ratio, or scaling factor, used in the mapping 
step, with all working shown.

(15 marks)

• Load an image


• Display an image


• Present choices to the user (menus or buttons)

Can't seem to get my threshold method working. Is it alright for u to 
check my codes and comment on it? I looked at the lecture slides 
(pseudocode) and tried to do it. Here is my codes.. 

  public int threshold(int amt) {
      for (int i = 0; i < pixels.length; i++){
          if (amt >= 100) {
               amt = 255;    
          } else {
               amt = 0;
          }
     }
     return amt;   
  }
 
=================

Well, a few things mixed up here!

amt is the threshold value.

This is the value that all the pixel values will be checked against.

If a pixel value is less (<) than this, the new pixel value will be 0 (black).

If a pixel value is more than or equal to (>=) this, the new pixel value will 
be 255 (white).

You DO NOT check whether amt is less than the threshold value - amt 
is the threshold value!

Inside the loop, you need to get every pixel value from the pixels array:

pixels[i]

Then check this against the threshold value (amt, in this case).

Then, you set the pixel value, pixels[i], to either 0 or 255.

Also, DO NOT Hard-code the threshold value.

In this case, the threshold value is hard-coded at 100 (amt, which should be 
the threshold value, is passed in as an argument to the threshold method).

NOTE ON COLOUR:

For your project, all original images can be greyscale.

This means that I don't care if your project doesn't work for colour images.

ie. Sir, should my project work for colour images?
    I don't care!

HOWEVER, one operation MUST work for colour - the operation to display only 
the RED, GREEN, or BLUE components of an image.

All the other operations can be for grayscale images only.

You may choose to do for colour, also - but not essential.

To threshold a colour image, you can extract the R, G, B component values,
then take the average of these.

Then, check the average against the threshold value - then set all component
values to 0 or 255.

Thanks for the question.

Keep 'em coming!

:-)

ps. I'm serious about TWO things regarding your project:

1. COPYING. I will be very STRICT about copying. ZERO marks (effectively fail 
MMPG) for copying.

2. NOT DOING YOUR OWN WORK. I know of one student who got a friend to do most 
of her ITEN project for her (calculations, etc). I will NOT tolerate this, 
and will question any student(s) I suspect. 

pps. To the student who left the rude comment (deleted) on this blog, please 
realise that a polytechnic diploma isn't so tough. If you can't handle it, 
take a good hard look in the mirror.