This research explores a novel method to enhance the security of graphical PIN-based passwords by leveraging celebrity recognition within Kanpur, Uttar Pradesh. By using familiar celebrity faces instead of textual passwords, the study aims to address issues of weak password selection and susceptibility to attacks. A recognition-based interview protocol identifies widely known celebrities across domains like sports, entertainment, and music, ensuring diversity and balance. Popularity scores, derived from frequency and order of mention, are analysed using statistical measures such as mean, standard deviation, and entropy to create balanced celebrity grids. This locally informed approach enhances both security and usability, offering a scalable framework for developing more secure, user-friendly authentication systems across different regions.