Real-Time Face Recognition with Mask using Deep Convolutional Neural Network

Proceedings of the 2023 4th International Conference on Computing, Networks and Internet of Things Pub Date : 2023-05-26 DOI:10.1145/3603781.3603863

Md. Ashif Mahmud Joy, Md. Fuad Hasan Khan Chowdhury, Sinha Afroz, Md. Nurul Islam, Ruaida Muhsinat, Mukta Akanda Moly, D. Farid

{"title":"Real-Time Face Recognition with Mask using Deep Convolutional Neural Network","authors":"Md. Ashif Mahmud Joy, Md. Fuad Hasan Khan Chowdhury, Sinha Afroz, Md. Nurul Islam, Ruaida Muhsinat, Mukta Akanda Moly, D. Farid","doi":"10.1145/3603781.3603863","DOIUrl":null,"url":null,"abstract":"The COVID-19 pandemic started in 2019, from this situation people learned that the use of face masks is one of the most effective ways to protect themselves from Coronavirus. A problem has arisen from this situation. Face recognition systems are widely used nowadays but all those systems are trained to detect perfectly exposed faces, not masked or occluded faces. As most people wear masks recently, it has become challenging for the existing face recognition systems to recognise faces. To suppress this problem, a feasible method for masked face recognition is proposed in the paper. For extracting the facial features of the non-occluded part of the face, VGG Face model is used. After extracting the facial features, those would be included in the dataset along with zoomed and rotated facial images for training. After that CS classifier is used for the classification and determines if the masked face is recognised or not. We have created Masked and Non-masked Face Dataset for the experiments.","PeriodicalId":391180,"journal":{"name":"Proceedings of the 2023 4th International Conference on Computing, Networks and Internet of Things","volume":"14 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2023-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 2023 4th International Conference on Computing, Networks and Internet of Things","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3603781.3603863","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

The COVID-19 pandemic started in 2019, from this situation people learned that the use of face masks is one of the most effective ways to protect themselves from Coronavirus. A problem has arisen from this situation. Face recognition systems are widely used nowadays but all those systems are trained to detect perfectly exposed faces, not masked or occluded faces. As most people wear masks recently, it has become challenging for the existing face recognition systems to recognise faces. To suppress this problem, a feasible method for masked face recognition is proposed in the paper. For extracting the facial features of the non-occluded part of the face, VGG Face model is used. After extracting the facial features, those would be included in the dataset along with zoomed and rotated facial images for training. After that CS classifier is used for the classification and determines if the masked face is recognised or not. We have created Masked and Non-masked Face Dataset for the experiments.

查看原文本刊更多论文

COVID-19大流行始于2019年，从这种情况下，人们了解到使用口罩是保护自己免受冠状病毒感染的最有效方法之一。这种情况产生了一个问题。人脸识别系统目前被广泛使用，但所有这些系统都被训练为检测完全暴露的人脸，而不是被遮挡或遮挡的人脸。由于现在大多数人都戴着口罩，现有的人脸识别系统对人脸的识别变得很有挑战性。为了抑制这一问题，本文提出了一种可行的掩码人脸识别方法。对于人脸非遮挡部分的人脸特征提取，使用VGG face模型。在提取面部特征后，这些特征将与缩放和旋转的面部图像一起包含在数据集中进行训练。然后使用CS分类器进行分类，判断被遮挡的人脸是否被识别。我们已经为实验创建了遮罩和非遮罩的人脸数据集。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Proceedings of the 2023 4th International Conference on Computing, Networks and Internet of Things

自引率

0.00%

发文量