With the rise of digital technology online accounts and sensitive data storage are top priorities. Although traditional text-based passwords have been around for a while now, many security breaches happen because of their weaknesses. Graphical Password Authentication, as an innovative alternative which uses the cognitive strengths of human memory and utilizes graphics as interactive elements to satisfy this restriction. By using images with dynamic refresh, shuffle and personalized patterns it suggests a more secure and user-friendly way to authenticate. The implementation of GPAuth, which uses graphical passwords—where a user chooses images in sequence as seen in GUIs. Unlike previous graphical passwords (e.g., pass points and persuasive click point) that are afflicted with the “hotspot” problem, GPAuth is resistant to over-the-shoulder attacks due to its unique features. In this paper, we propose a system which incorporates features like Random Image Shuffling, used for shuffling images in the set, Image refreshment which changes the set of images in a particular time interval and, malicious activity detection, for preventing unauthorized access using techniques like wrong sequences, two-factor authentication, trial attempt, etc. Taking into consideration above features, GPAuth excels the traditional systems by observable difference in security, usability, and efficiency. Finally, we present this system, discussing its design and implementation while shedding light on the potential implications for digital security.