Anne Dawson: CSCI101A_FA_FA04.htm Friday 29th October 2004, 11:28 PT, AHD

This document is subject to change without notice.

Please report any errors or omissions in this document to:

adawson@coquitlamcollege.com

Special instructions: For this assignment you may work in teams of 2, or alone.

This assignment is due at the start of the last class of the semester (Week 13 Class 2).

Final Assignment Specification

------------------------------

Introduction

------------

A digital image can be represented as a collection of small square regions known as pixels -the word pixel is short for "picture elements". In the case of a colour image, each pixel is represented by a red, green and blue triple. Each number in the triple ranges from 0 to 255, where 0 indicates that none of that primary colour is present in the pixel and 255 indicates that a maximum amount of that primary colour. This means that an RGB of 0 0 0 represents a complete absence of colour (i.e. black) and 255 255 255 is the opposite (i.e. white). Other examples: 255 0 0 represents red, 0 255 0 represents green and 0 0 255 represents blue.

An image is made up of a grid of rectangular areas (pixels) of the displaying medium (a computer screen, printed paper etc.), where each rectangle has one colour represented by an RGB triple. The following grid represents 9 pixels where the first pixel is white:

255 255 255 142 160 210 142 160 210

136 158 208 135 158 209 136 156 207

140 153 206 142 151 206 142 151 206

Image Files

Image files contain a map of the image as a collection of bits. They are known as bit map files and have the file extension .bmp. There are other types of graphic bit map files which can store the same image, but in different compression formats (.jpg, .gif). The Unix (or Linux) utility xv converts a bit map to the equivalent ASCII file. Hence the bit map 11111111 is converted to ASCII 255 which is the decimal equivalent of binary 11111111.

The following is an example of a converted bit map image:

255 255 255 142 160 210 142 160 210

136 158 208 135 158 209 136 156 207

140 153 206 142 151 206 142 151 206

Image Processing

----------------

Changing the hue of an image:

It's possible to modify an image by manipulating the pixel values. For instance, if you want to make the image more red, you would increase the value of the red component of the RGB triple.

For example, a redder version of the image pattern above would be this:

255 255 255 255 160 210 255 160 210

255 158 208 255 158 209 255 156 207

255 153 206 255 151 206 255 151 206

Smoothing the image:

The process known as smoothing results in a blurred version of the original image:

Original image:

Smoothed image:

Smoothing is a process which replaces the value of every pixel with an average value of the neighbouring pixels.

Example

A two-dimensional array of pixels represents our image. Each pixel has three numeric values that represents the amount of red, green and blue. To perform the smoothing process on the image, we will take an average of the current pixel and the four adjacent pixels. This requires taking an average of each of the three colours for each pixel.

Consider the following:

243 159 211 142 160 210 142 160 210

136 158 208 135 158 208 136 156 207

140 153 206 142 151 206 142 151 206

The current pixel has a red value of 135, a green value of 158, and a blue value of 208.

The smoothed value for this pixel is calculated as follows:

red value -> (135 + 136 + 142 + 136 + 142)/5 = 138,

green value -> (158 + 158 + 151 + 156 + 160 )/5 = 156,

blue value -> (208 + 208 + 206 + 207 + 210)/5 = 207,

Notice that the West, South, East and North pixel value of the current pixel are summed to the

current pixel value. For example, the pixels averaged to calculate the red value are:

243 159 211 142 160 210 142 160 210

136 158 208 135 158 208 136 156 207

140 153 206 142 151 206 142 151 206

This process is repeated for every interior pixel in the array (one which is surrounded on each side by pixels).

Program Specification:

Your program should read an ASCII graphics file, and provide the options to smooth the image, and to make the image more red, green or blue. The resulting image must be sent to an output file.

For maximum points, your program should work with a file containing any size of 2 dimensional grid of RGB decimal triples.

You must store the RGB triple in a structure defined as follows:

struct pixel

{

unsigned int red;

unsigned int green;

unsigned int blue;

}

and you must use a two-dimensional array of these structures in your program.

This document is subject to change without notice.

Submission instructions

-----------------------

At the start of class (Week 13 Class 2) you should save just your source code file to your folder in \Week13\FA.

If you are working in a team, both team members save the same file to their own folder.

Marking Scheme

--------------

The following marking scheme applies:

Course Code: CSCI101A

Semester: FA04

Assignment Code: Final Assignment

Lab Specification: Image Processing (img.cpp)

Instructor Name: Dr Anne Dawson

Student 1 Name:

Student 1 Number:

Student 2 Name:

Student 2 Number:

DESIGN

1. The program has appropriate modularity i.e.

functions are used where it makes sense to

use them. /10

2. Appropriate data types and control structures

are used. /10

3. The program is robust

(handles exceptional circumstances). /10

4. The program is efficient

(does not contain unnecessary statements) /10

MAINTAINABILITY

5. The program is commented appropriately - including

date, filename and pre- and post-condition

comments. /10

6. The program has meaningful identifiers. /10

7. The program is indented (spaced out) correctly, to

aid the understanding of the code. /10

8. The code is easy to follow. /10

CORRECTNESS:

9. The program runs as intended (is accurate). /10

10. Comprehensive test data and results are supplied. /10

% Complete:

Bonus:

Total: /100

Date: