Abstract

The main objective of this research is to create a more secure and user-friendly password-generation method that is less vulnerable to attacks such as shoulder surfing. Similarly, while textual or alphanumeric passwords are not entirely secure, they also come with certain limitations, including issues related to memory. The novelty of this research is the development of a new algorithm (Secure Password Algorithm/SPA) for password generation that uses graphics to generate textual input from users based on user perceptions. This method is more resilient against attacks like shoulder surfing and offers a more secure and user-friendly way of creating passwords. This research uses a combination of literature review and SPA proposals to create passwords with graphical and encryption techniques. The literature review examines the weaknesses of existing graphical and text password methods. The proposed algorithm focuses on user perception and confidential data to create more secure and easy-to-use passwords. The research results show that this research can provide a new, more secure, and user-friendly method of generating passwords such as shoulder surfing. Research supports software implementation of these algorithms and psychometric, psychological, and psychiatric studies to improve user input and protect systems from malicious attacks. Graphical passwords provide better recall compared to textual passwords, and the proposed algorithm for password generation using graphical and encryption techniques provides a more secure and user-friendly method of password generation that is less vulnerable to attacks such as shoulder surfing. The proposed new encryption algorithm has succeeded in overcoming various potential weaknesses and attacks on previous encryption methods. The test results demonstrate the efficiency and near invulnerability of the SPA algorithm against brute force attacks on specific security parameters. Tested alongside AES, shows comparable security levels, particularly in resource-constrained environments. To address doubts, the paper establishes evaluation standards for computational complexity, resistance to known attacks, and adaptability to evolving threats. The algorithm has the potential to be a stronger and more secure solution for generating passwords to safeguard confidential user information.