http://www.iaeme.com/IJMET/index.asp 501 editor@iaeme.com International Journal of Mechanical Engineering and Technology (IJMET) Volume 10, Issue 02, February 2019, pp. 501–510, Article ID: IJMET_10_02_051 Available online at http://www.iaeme.com/ijmet/issues.asp?JType=IJMET&VType=10&IType=02 ISSN Print: 0976-6340 and ISSN Online: 0976-6359 © IAEME Publication Scopus Indexed ILLUMINATION ENHANCED FACE RECOGNITION FOR SMART ACCESS SYSTEM Anusha Hegde Department of ICT, Manipal Institute of Technology, MAHE, Manipal Prathviraj N and Nishmitha R. Shetty Department of Computer Science & Engineering, St. Joseph Engineering College Mangaluru ABSTRACT. Principle Component Analysis (PCA) is a widely used technique in the field of face recognition. Any system will have its own pros and cons. One of the disadvantages of PCA is its bonding with the light variation. By focusing on this issue, an improved algorithm is proposed by combining PCA with the Fast Fourier Transofrm-2. This solution is then applied to a real time environment which provides the smart access to the house. The user interface is provided using the Android application installed on the user’s smart phone. The communication between the smart access system and the application can happen remotely. The mechanical door lock is controlled using the system which can receive and transmit the command. Keywords: PCA, Face Recognition, Aurdino Uno, IoT, Smart Door Lock Cite this Article: Anusha Hegde, Prathviraj N and Nishmitha R. Shetty, Illumination Enhanced Face Recognition for Smart Access System, International Journal of Mechanical Engineering and Technology, 10(02), 2019, pp. 501–510 http://www.iaeme.com/IJMET/issues.asp?JType=IJMET&VType=10&IType=02 1. INTRODUCTION Recognition of face is commonly used for identifying a person because it has the capability to calculate and recognize the human identity, which is ideal for security purposes. It is proved to be the most advantageous biometric strategy among the others. The study of inherently large- dimensional essence of the training data in a lower dimensional manifold has become popular in recent decades. This is known as Dimensionality Reduction. The technique that is of interest in this work is based on a Linear Subspace method. Principal component analysis (PCA) is one of the commonly used linear subspace method for face recognition. It is a statistical process that makes use of an orthogonal transformation for altering the set of observations for corresponding data to a set of values of linearly uncorrelated data knows as principal components. PCA was proposed by Kirby and Sirovich to implement karhuncn-Ioevc (K-L) transform in order to represent human face as a linear combination of weight vectors. Later, Turk et.al [1] developed face recognition technique using PCA (KL transform) for subspace method.