Faculty Advisor

Agu, Emmanuel O.

Abstract

The goal of this project was to extract key features from photographs of faces and use machine learning to classify subjects as either sober or drunk. To do this we analyzed photographs of 53 subjects after drinking wine and extracted key features which we used to classify drunkenness. We used random forest machine learning to achieve 81% accuracy. We built an android application that using our classifiers to estimate the subject’s drunkenness from a selfie.

Publisher

Worcester Polytechnic Institute

Date Accepted

April 2018

Major

Computer Science

Project Type

Major Qualifying Project

Accessibility

Unrestricted

Advisor Department

Computer Science

Share

COinS