Javascript must be enabled for the correct page display

Social Density Estimation Based on Consumer Smartphone Sensors

Putra, G. D. (2017) Social Density Estimation Based on Consumer Smartphone Sensors. Master's Thesis / Essay, Computing Science.

[img]
Preview
Text
thesis-guntur.pdf - Published Version

Download (24MB) | Preview
[img] Text
Toestemming.pdf - Other
Restricted to Backend only

Download (77kB)

Abstract

Recent developments in smartphone technologies raise the concept of mobile healthcare systems as an essential part of medical care or research processes. As opposed to the conventional techniques which are prone to human errors and biased results, smartphone based monitoring systems can provide objective results especially when dealing with longitudinal assessment of individual movement patterns in the context of social density. In this thesis, we present a consumer smartphone based social density estimation method that estimates the number of people in a certain area by utilizing smartphone sensors. We use WiFi to count nearby Access Points (AP) and microphones to record ambient noise of the surroundings. We performed data collection in several locations, ranging from low to high level social densities, using WiFi MAC address counting and time-lapse images as the ground truth approximation. The results indicate that smartphones have good potential for estimating social density levels. The results reveal that the AP and ambient noise have a positive correlation with the social density level, which in our experiments is 0.8 and 0.6, respectively. Furthermore, we also constructed prediction models for the social density level using new data with residual error is equal to 7.05.

Item Type: Thesis (Master's Thesis / Essay)
Degree programme: Computing Science
Thesis type: Master's Thesis / Essay
Language: English
Date Deposited: 15 Feb 2018 08:26
Last Modified: 15 Feb 2018 08:26
URI: http://fse.studenttheses.ub.rug.nl/id/eprint/14880

Actions (login required)

View Item View Item