APPLICATION OF HAAR CASCADE METHOD IN MASK DETECTION APPLICATION

  • Galang Aprilian Anarki Teknik Informatika, Institut Teknologi Nasional Malang
  • Karina Auliasari Informatics Engineering, National Institute of Technology in Malang
  • Mira Orisa Informatics Engineering, National Institute of Technology in Malang
DOI: https://doi.org/10.36040/jati.v5i1.3214
Keywords: haar cascade, mouth detection, mask detection, object detection, python, covid-19

Abstract

This research aims to create an application that can detect the use of masks to minimize the transmission of the Covid-19 virus, which is currently an epidemic in Indonesia, by issuing an audio warning feature and taking pictures if a mask is not detected.
This study using the Haar Cascade method. Haar Cascade is an object detection method created by Paul Viola and Michael Jones. In 2001, they presented a paper called "Rapid Object Detection using a Boosted Cascade of Simple".
The results of this study are the application can detect masks from images sourced from photos or videos from internal and external webcams well, with the highest total accuracy of 88.7% and the lowest 44.9%. The warning features in the form of audio and shooting can also work well.

Downloads

Download data is not yet available.
Published
2021-02-27
Issue
Vol. 5 No. 1 (2021): JATI Vol. 5 No. 1
Section
Articles