LIGHT-WEIGHT FACE MASK DETECTION USING CENTER NET DEEP LEARNING DETECTOR
- Asiya1, C. Shoba Bindu2, P. Dileep Kumar Reddy3
1Department of CSE, JNTUA College of Engineering, Ananthapuramu, Andhra Pradesh, India, email@example.com
2Department of CSE, JNTUA College of Engineering, Ananthapuramu, Andhra Pradesh, India, firstname.lastname@example.org
3Department of CSE, Narsimha Reddy Engineering College (Autonomous), Secunderabad,Telangana, email@example.com
The Coronavirus disease had a significant effect on the world. One significant defensive intervention for people is to put on veils in open provinces. A few districts applied a necessary veil wearing principle in open regions to forestall transmission of the infection. Hardly any examination studies have inspected programmed facial covering discovery in light of picture examination. In Covid-19 circumstance is significant thing we ought to safeguard ourselves by wearing covers in open regions. A few districts gave obligatory cover – wearing standard in open regions to keep transmission of the infection starting with one individual then onto the next individual. In our paper, by using CenterNet Deep Learning Detector, Figuring out if someone is using a mask or not. A Red colored rectangular box is accurately formed all around face of any individual in the web camera who is not using a surgical mask/face mask and a Green colored rectangular box is accurately painted all around face of anyone who is donning a mask. Our CenterNet model achieved 99% validation accuracy, 98% testing accuracy, 99% F1-Score, 99% precision accuracy and 99% recall accuracy.
INDEX TERMS: YOLO V3 (You Only Look Once), Face Mask Detection, CenterNet, Deep Learning(DL) Detector, COCO dataset,