Face Mask Detection using Gabor Filter and Gamma Transform with Localized Feature Extraction

The COVID-19 pandemic has necessitated widespread adoption of face masks as a preventive measure against virus transmission. To address the challenge of detecting face masks accurately and efficiently, this paper introduces a novel approach that combines Gabor Filters with Gamma Transformation. The proposed method aims to enhance feature extraction and classification accuracy in masked face recognition. Through comprehensive experimentation on public medical face mask datasets, the approach demonstrates a significant improvement in accuracy compared to conventional methods. By leveraging the synergy between Gamma Transformation and Gabor Filters, The gamma-transformed facial image undergoes deconstruction using Gabor filters. Leveraging wavelet analysis facilitates the localization of image features in both space-frequency domains, offering diverse spatial resolutions and orientations. the proposed method achieves an accuracy enhancement from 86.32% to 94.78%. Experimental results highlight the effectiveness of the proposed approach in accurately identifying masked faces, surpassing the performance of conventional techniques across diverse datasets.