Catalog Home Page

Development of 3D image manipulation software utilizing the microsoft kinect

Ladzinski, Aston (2012) Development of 3D image manipulation software utilizing the microsoft kinect. Internship Report, Murdoch University.

[img]
Preview
PDF - Front Pages
Download (675kB)
[img]
Preview
PDF - Whole Thesis
Download (1MB)

Abstract

Studying Engineering at Murdoch University enables students to experience many facets of engineering. In the Electrical Engineering based courses there is a vital need to have an understanding of various programming languages and methods and then explore ways in which this knowledge can be utilised. This document describes how the Microsoft Kinect can be utilized to control 3D images, specifically medical images, through the use of programming skills and software development kits.

This project involves learning how the Microsoft Kinect sensor actually works and requires the development of two programs that utilised this sensor and can be easily implemented. The first program is designed to display information that the Kinect sensor is able to detect. By developing this program, it enables a user to quickly gain an understanding of what data is available for manipulation. The second program is designed to manipulate a generic 3D image through the use of a set of gestures initiated by the user. This enables the user to see how the information retrieved from the sensor in the first program can be manipulated for useful purposes, by the creation of algorithms.

The development of these two programs required the learning of a new language. The language that was used is called C#, and it is a complex object orientated language. Through the use of the language in the Microsoft Visual Studios 2010 programming environment, it was possible to create the programs through many iterations of development. The thesis documents how this approach was made, both the method of learning C# and also incorporating and utilising the benefits of C# into the programs.

From the research conducted it was found that creating effective gestures was quite difficult as it required some form of predictive logic, and that incorporating medical 3D imagery was at a completely different level of programming skills. Despite these difficulties an effective, easy to use and modifiable program was developed that will allow future research to continue in this field. Therefore possible future projects and developments are also discussed in order to give the reader an idea of what the sensor is capable of if time and knowledge are readily available.

Publication Type: Internship Report (Bachelor of Engineering)
Murdoch Affiliation: School of Engineering and Energy
URI: http://researchrepository.murdoch.edu.au/id/eprint/13111
Item Control Page Item Control Page

Downloads

Downloads per month over past year