International Journal of Computer and Information Technology (ISSN: 2279 – 0764) Volume 12– Issue 02, June 2023 www.ijcit.com 79 Identification of Medical Mask Use by Applying the Convolutional Neural Network Algorithm and the Gabor Filter with Multiclass Classification Muh Arifandi University of Technology Yogyakarta: Master of Information Technology Yogyakarta, Indonesia Email: muh.arifandi [AT] student.uty.ac.id Erik Iman Heri Ujianto University of Technology Yogyakarta: Master of Information Technology Yogyakarta, Indonesia Email: erik.iman [AT] uty.ac.id Abstract— Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) causes global pandemics and makes countries around the world lock down for tourists. This action is required to prevent the spread of viruses that take 14 days to disappear. SARS-COV-2 can easily infect individuals through a droplet. Thus, the governments of every country worldwide recommend wearing medical masks to prevent the spread of viruses, as well as maintaining distance during activities with others and washing hands frequently. Medical masks become efficient if their application is precise, owing to a lack of knowledge and self- awareness to preserve their distance and wash their hands. This paper proposes a Convolutional Neural Network (CNN) with Gabor filter implementation. The simulation uses a mask on a dataset with over 70,000 individual photos. The results demonstrated that the proposed CNN-Gabor model in this work could effectively classify the position of the mask when compared to the CNN model without the Gabor filter. Keywords-convolutional neural network, gabor filter, medical mask classification, multiclass classification. I. INTRODUCTION The world is experiencing a pandemic caused by Severe Acute Respiratory Syndrome Coronavirus 2. (SARS- COV-2). The existence of this pandemic prompts several countries to undertake lockdown measures in order to prevent the virus spread and the strain on health workers [1]. Patients infected with Covid-19 require around 14 days to heal or complete the isolation period before it is definitively confirmed that they no longer transmit the virus. Based on epidemiological studies and virology proves that Covid-19 is mainly transmitted from people who are symptomatic to others who are in close range through droplet [2]. Transmission of droplets occurs when someone is at a close distance that is about 1 meter from someone who has breathing symptoms such as coughing or sneezing, so the droplet is at risk of sticking to other people. Based on these conditions, COVID-19 prevention includes wearing medical masks, keeping a distance in activities, routine hand washing, and constantly applying health protocols. Machine Learning (ML) is a branch of artificial intelligence in which a machine uses intelligent software to solve problems [3]. Tensorflow is one of many libraries that are collections of functions that can be utilized to develop ML algorithms [4]. ML can carry out specific tasks by studying data and statistical models used during training. In general, ML can be divided into three types: (1) supervised learning, (2) unsupervised learning, and (3) deep learning (DL). DL is a subset of Artificial Intelligence (AI), including a hidden layer that processes RAW datasets [5]. DL employs a simple representation yet can build complex concepts based on the dataset used during the training process. Several algorithms can be used to train models, such as mobile networks, VGGnet, and so forth. Following that, object detection can use algorithm types such as (1) Yolo, (2) SSD Resnet, and (3) MTCNN. II. RELATED WORK Several studies have been done to limit the spread of the Covid-19 virus, including research into the use of artificial intelligence in the pandemic era, such as mask detection. Ejaz conducted one of the studies related to introducing the face of a mask to classify the use of masks in the scope of society [6]. The detection value based on accuracy in the study ranged from 98% to 99%. Venkateswarlu conducted similar research in which the authors employed the mobilenet model.