Pixel Value Graphical Password Scheme: Compatibility of K-means Clustering Algorithm as Pixel Value Password Fault Tolerance Mechanism

Mohd Afizi Mohd Shukran,Mohd Sidek Fadhil Mohd Yunus,Mohammad Adib Khairuddin,Norshahriah Abdul Wahab,Mohd Nazri Ismail,Kamaruzaman Maskat,Muhammad Naim Abdullah,Mohd Fahmi Mohamad Amran,Mohd Rizal Mohd Isa

doi:10.11648/j.ijsmit.20190502.13

Mohd Afizi Mohd Shukran, Mohd Sidek Fadhil Mohd Yunus + Show 7 more

Open Access

https://doi.org/10.11648/j.ijsmit.20190502.13

Copy DOI

Abstract

In September 2018, the patent for pixel value graphical password scheme was granted in Malaysia. The graphical password scheme was designed to reduce the complexity of previously developed graphical password scheme where a user only requires to load their personal image as password instead of complex graphical challenge during authentication. As the guardian of digital access, Pixel Value Access Control was highly invincible from password pixel forgery attack where a little bit different pixel value derived from loaded image will deny the access. Only the original enrolled image from a registered user can be recognized by Pixel Value Access Control to authenticate the respective username. That fact makes the graphical password scheme is a trusted access control mechanism but, on the other hand, it makes users bound with the only original password pixel image file. Thus, Pixel Value Access Control need to be installed the pixel value fault tolerance mechanism where it could allow users to acquire their password pixel image file from various storage media. The clustering technique was suggested to solve this issue where it allows an altered pixel password being recognized as the same group of the original pixel password. There are number of clustering algorithms developed for various purposed and application of digital image clustering. K-Means algorithm is one the partition-based clustering algorithm that found to be the simplest and fastest clustering algorithm as suggested by many researchers. This paper is mainly to exhibit the selection of K-Means clustering algorithm became the crucial algorithm for Pixel Value Access Control password pixel fault tolerance algorithm. Background of this topic was briefly explained in introduction section, the implementation of K-Means algorithm as Pixel Value Access Control fault tolerance was elaborate in section 2 and followed by validation of the implementation in section 3. At the end of this paper, there is conclusion for this study and followed by list of references.

Highlights

The Pixel Value Access Control or shortly called as PVAC, [1] is an access control system that developed based on pixel value graphical password scheme method where it extract digital image file to validate a username
With Pixel Value Access Control, only users could use their credential image while others became inaccessible where the identity key is only hold by the actual users
The username authentication key is in the form as a digital image file that originates from users personal image storage provide by users during the enrollment and authentication process through the access control interface

Summary

Introduction

The Pixel Value Access Control or shortly called as PVAC, [1] is an access control system that developed based on pixel value graphical password scheme method where it extract digital image file to validate a username. As the digital image files hold a pixel value, Pixel Value Access Control extracts the pixel value as the password pixel, PassPix. There are 3 type of feature recognition for clustering techniques which are: i. Cloud storage media conditions and settings are uncertain where some of them are compressed and caused the pixel value alteration on PassPix. One of the common practices of cloud file repository is to transferred and stored files in WhatsApp [13, 3] messaging application where users could send, share and save their files in the app’s cloud without draining their machine storage disk. Adapting a clustering technique into Pixel Value Access Control is a possible way as fault tolerance mechanism where the altered PassPix is recognized as the same category

The Clustering Techniques

Validation of K-Means Compatibility for Pixel Value Access Control

Conclusion