Please use this identifier to cite or link to this item:
http://dspace.dtu.ac.in:8080/jspui/handle/repository/20405
Title: | FACE MASK DETECTION USING CONVOLUTION NEURAL NETWORK BASED MACHINE LEARNING MODEL |
Authors: | SUNNY |
Keywords: | FACE MASK DETECTION CONVOLUTION NEURAL NETWORK MACHINE LEARNING MODEL |
Issue Date: | May-2023 |
Series/Report no.: | TD-6873; |
Abstract: | In the year 2020 the world witnessed a global health emergency due to the outbreak of virus pandemic Coronavirus named COVID-19 also the increasing of the air polluƟon in the ciƟes of developing countries due to emission of PM10 and PM2.5 pollutants from burning of Organic Fuels ,fast urbanizaƟon the usage of face masks in public has now become a way of life .The usage of face mask is not only mandated by the governments to control the spread of the virus but also a recommendaƟon made by doctors to protect lungs of paƟents. Due to the usage of facial mask in public spaces with huge amount of fooƞalls namely markets, shopping malls, public transport like metro rails, sporƟng events ,music concerts manually check the proper usage of masks is not only a tedious and difficult task but also an impossible one for big country like India where populaƟon density is one of the highest in the world. The Public Monitoring systems widely used face many challenges to correctly monitor the usage of face mask due to difference in the mask types, low quality Cameras, ObsfuscaƟon of faces etc. Also, majorly the lack of huge amount of data to be trained on is one of the main challenges. Therefore, this project is aimed at providing a comprehensive review of exisƟng machine learning models that have been used to detect face masks and developing an ensemble approach for the same using a newer balanced dataset not widely used before. |
URI: | http://dspace.dtu.ac.in:8080/jspui/handle/repository/20405 |
Appears in Collections: | M.E./M.Tech. Computer Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
SUNNY M.Tech.pdf | 3.74 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.