Improving Smart Infrastructure Monitoring System as a Response to Prevalent Pandemic | ||||
| Port-Said Engineering Research Journal | ||||
| Volume 27, Issue 1, 2023, Page 24-30 PDF (439.67 K) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/pserj.2023.209431.1236 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Nourhan Osama Mohamed Mokhtar  1; Ahmed Magdy2; Basem El-Hady3; khaled abd elsalam4
| ||||
| 129Ard Alandia,Ismailia. | ||||
| 2Electrical Engineering Department, Faculty of Engineering, Suez University, Ismailia, Egypt | ||||
| 3Electrical Engineering Department, Faculty of Engineering, Suez Canal University, Ismailia, Egypt | ||||
| 4Electrical Engineering Department, Faculty of Engineering, Suez Canal University | ||||
| Abstract | ||||
| Face masks are no longer an option for protection against airborne diseases brought on by coughing, talking, or sneezing, which can spread germs into the air and infect everyone nearby, especially with the coronavirus pandemic that occurred in 2019. Also, some states and the government have made it mandatory for people to wear face masks. In recent years, Artificial Intelligence has played an important role in the medical field, as Convolution Neural Network techniques have proven to be very useful in image detection applications with different algorithms. In this paper, we propose a model using deep learning algorithms to achieve the most efficient and speedy way to detect the presence of a face mask on people in public places by using RGB cameras. The Alexnet, Googlenet, Resnet 18, and Squeezenet are trained on a dataset that consists of images of people with and without masks and is publicly available as “Mola RGB Covsurv” Mendeley Data, with 80% of the dataset being used for training and 20% for testing to get the most efficient algorithm. The proposal we recommend is Squeeznet's algorithm, which achieved an average precision of 94.1592% with a sensitivity of 91.19533% in 1700 minutes and 10 seconds | ||||
| Keywords | ||||
| Coronavirus; Artificial intelligence; Deep Learning; Convolution neural network | ||||
|
Statistics Article View: 89 PDF Download: 70 |
||||
