Project 1 : Image Processing and GUI in Java

Nik Tuzov, Ed Dudek

The goal of this project was to create an image-processing application with GUI. The project was developed in NetBeans IDE 5.0 (under Linux) and the following features are available:

•  Opening a directory and loading an .rgb image

•  Detecting the edges of the image, displaying the result and saving it to a hard drive

•  Ranking images based on their similarity to the currently open image and displaying the result

Below you can download:

  <--  NetBeans project folder (it has the README file with more information)

  <--  a report with a class diagram created in VP for UML

   <--  a few sample *.rgb files and a MatLab program to convert .jpg to .rgb

Project 3 : Multithreaded programming in Java

Nik Tuzov, Matthew Glause

The goal of this project was to create an application that would take advantage of multithreading. The project was developed in NetBeans IDE 5.0 (under Linux).

The program processes a specified directory of *.rgb images sharpening each one and saving it in the same directory. The pivotal point in this project is how to assign those *.rgb files to the threads to achieve a maximal speed-up. Our approach is to assign the next .rgb file to a thread that has been assigned the least number of pixels up to the current moment. After this assignment is done, all threads get started asynchronously (you can see that in the sequence diagram).

As a result, the average process time is about 13 microseconds/pixel with just one thread and 7 microseconds/pixel with two threads. It turned out that increasing the number  of threads from 2 to 8 doesn't  really speed it up - probably because the advantage of multithreading gets negated by the necessary overhead.

This application doesn't provide any tools to view the .rgb images, but you can do it with Project 1 above.

Below you can download:

 <--  NetBeans project folder (it has the README file with more information)

  <--  a report with  class and sequence diagrams created in VP for UML

  <--  a few sample *.rgb files and a MatLab program to convert .jpg to .rgb (the same files as for Project1)