Smart Shaving Mirror

Pitt, Peter (2020) Smart Shaving Mirror. [USQ Project]

[img]
Preview
Text (Project)
Pitt_P_Brown_Redacted.pdf

Download (926kB) | Preview

Abstract

Shaving of a face can be difficult for many people. This task can be made even more challenging for a person who encounters difficulty with either movement or vision. This project investigated the use of smart mirror technology to produce a mirror that would assist the user in the shaving process. The smart mirror was intended to intuitively provide an enlarged image of the area of the user’s face that is being shaved. The selected region would be determined by tracking the movement of the shaving device being used.

Based on the literature review, three tracking algorithms were produced and tested. Two of these were based on the K-means algorithm. The first used the traditional method of K-means clustering, while the second implemented two different optimisation techniques discovered in the literature.

The first technique reused the cluster centres from the previous frame. This technique lowered the number of iterations required to complete the clustering and therefore, the computational time decreased. The second method checked if the Euclidean distance from the pixel to the cluster had increased as a test to determine if the pixel needed to have its cluster assignment recalculated. The combination of these two techniques resulted in a time saving of over 60% when incremental differences were present in the frames.

The third tracking technique implemented was the cross-correlation algorithm. The cross-correlation algorithm in its original form is very heavy computationally. However, through the use of the integral image technique, the algorithm operated with high efficiency.

Implementing the software in a more efficient language, such as C++, has the potential for completing a working smart mirror with the desired capabilities.


Statistics for USQ ePrint 43036
Statistics for this ePrint Item
Item Type: USQ Project
Item Status: Live Archive
Additional Information: Bachelor of Engineering (Honours)(Electrical & Electronic Engineering)
Faculty/School / Institute/Centre: Current - Faculty of Health, Engineering and Sciences - School of Mechanical and Electrical Engineering (1 Jul 2013 -)
Faculty/School / Institute/Centre: Current - Faculty of Health, Engineering and Sciences - School of Mechanical and Electrical Engineering (1 Jul 2013 -)
Supervisors: Brown, Jason
Date Deposited: 12 Aug 2021 01:50
Last Modified: 12 Aug 2021 01:50
URI: http://eprints.usq.edu.au/id/eprint/43036

Actions (login required)

View Item Archive Repository Staff Only