Diagonally Assisted DCT Technique for Face Recognition: DA-DCT
Author(s) -
Nipun Behl,
Shivangi Katiyar
Publication year - 2016
Publication title -
international journal of computer applications
Language(s) - English
Resource type - Journals
ISSN - 0975-8887
DOI - 10.5120/ijca2016909848
Subject(s) - computer science , discrete cosine transform , face (sociological concept) , artificial intelligence , diagonal , computer vision , pattern recognition (psychology) , image (mathematics) , mathematics , geometry , philosophy , linguistics
is a multidimensional intricate structure that demands good computing technique for recalling it. This paper proposes a new diagonally assisted face recognition technique using Discrete Cosine Transform (DCT). Proposed work divides an image into matrix to apply DCT. Working scheme i.e. DA-DCT selects diagonal coefficient to retain low, moderate and high frequency coefficients for each block. Proficiency of the proposed work has been checked for self- created database using MATLAB. Experimental results prove that DA-DCT performs much better than PCA. KeywordsCosine Transform, Face recognition
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom