Abstract: Mask Detection is commonly used as a pre-processing step before face recognition and multi-angle problemsbecause face recognition algorithms require a front face in the input image. COVID-19 survivors, on the other hand, protectthemselves by wearing face masks that cover the majority of their faces. As a result, under the current circumstance, someclassic methods of determining head locations are no longer valid. The HGL methodology for categorizing head postureswas established as a result of this research, which involved analyzing the color texture of photographs and line portraits.The suggested HGL technique trains the CNN to extract features for classification by combining the H channel of the HSVcolor space with a gray scale image of a facial portrait. When evaluated on the MAFA dataset, the proposed strategyoutperforms techniques based on facial landmark detection and convolutional neural networks.Keywords: - Social Media, Data Mining, Social Networks, Counselling, Educational Application, EaaS.