Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{196332,
        author = {Syeda Sayeeda Farhath and P. Naga Bhargav Reddy and Siddharth Karanam and Baalne Anjali},
        title = {Face Mask Detection in Public Health Using YOLOv8 with Explainable Artificial Intelligence},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {16258-16266},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=196332},
        abstract = {In highly populated locations where public health regulations must be promptly and efficiently enforced, there is an increasing need for automated compliance verification systems. During the COVID-19 outbreak, wearing a face covering was crucial for preventing the spread of respiratory droplets. In vast public spaces, manual compliance checks are labour-intensive and ineffective on a broad scale. Vision-based automated solutions are an excellent choice since they make use of current camera networks to identify compliance patterns. Explainable Artificial Intelligence (XAI) components are added to the YOLOv8 object recognition architecture to create an intelligent detection framework. In order to make automated surveillance more transparent, our approach combines fast object localisation with approaches to make things easier to grasp. To demonstrate which areas of a picture are in charge of the model's classifications, we employ Gradient-weighted Class Activation Mapping (GradCAM) and Local Interpretable Model-Agnostic Explanations (LIME). Using a publically accessible dataset of face masks, we train our system and evaluate its performance using common object detection measures such as accuracy, recall, F1-score, and mean Average accuracy (mAP). The findings demonstrate that our approach can draw conclusions in real time while still offering dependable detection accuracy. The explainability visualisations demonstrate that the model concentrates on semantically significant face regions, particularly the mouth and nose, where mask coverage is most crucial. Explainability characteristics increase public trust in AI-assisted decision systems and encourage more people to adopt machine learning-powered public health monitoring technologies.},
        keywords = {},
        month = {April},
        }