Advanced security system in video surveillance for COVID-19

Abstract

Image processing techniques are frequently used to scale, resize, compare, and transform graphical contents. The detection of face mask pattern from an image set could be frequently made using the Convolutional Neural Network (CNN) model. But during CNN based classification, several issues are observed. There has been limited work to improve the performance of face mask detection during COVID–19. However, there is much research in image processing but the time taken for prediction needs to be reduced. Moreover, there is the issue of space consumption by graphical content. Proposed research is supposed to minimize the prediction time and space consumption. Research has focused on the study of existing image processing research and techniques and eliminating their limitations. Research proposes a methodology for face mask detection using an edge–based convolution neural network algorithm. The elimination of useless content from a graphical image before applying Convolutional Neural Network has reduced time consumption. Moreover, it has also reduced the storage requirement for the graphical dataset. As the number of data sets increases, every comparison creates a huge gap in size and comparison time. Proposed work is supposed to implement the proposed methodology using MATLAB. Comparison of the proposed methodology and algorithm with the traditional algorithm is made during simulation. The proposed work is found to be more efficient compared to traditional techniques used in face mask detection. The use of this proposed work in COVID–19 is supposed to improve the capability of convolution neural networking at the time of decision making. The proposed work is supposed to be more accurate compared to the traditional mode. The proposed work would integrate the CNN approach with edge detection mechanisms to improve the performance of the face mask detection mechanism. These systems are connected to a smart video surveillance system to support the notification system in case of absence of a mask. If a mask pattern is not found, then the alarm is triggered to represent the presence of a person without a mask.