Image Transformation Based Hybrid Graphical Password Authentication System

Abstract

Authentication establishes identity and it is a process of proving the correctness of association between credentials and the owner. Now a day, the majority of applications requires authentication. Text based passwords are commonly used for authentication. However, providing security through text based passwords is somewhat undesirable. Graphical password authentication (GPA) is an alternative way of replacing text passwords to enhance security. GPA is motivated by the fact that images are more memorable than the text password. This paper proposes a new hybrid graphical authentication model focussing on image click based and text password selection by using the concept of image transformations. For each image, three types of transformations such as normal image, mirror image and shift image are considered in the proposed model. Even for the password also these three types of transformations are applied based on the image transformation which is displayed at the time of login session. The aim of developing the proposed model is to provide both security and usability by combining the principle of image transformation and change of password based on the transformed image. The radical change of password for each transformed image strengthens the security of the proposed hybrid graphical authentication system and also it provides resistance to shoulder surfing attack, guessing attacks, etc.