Paper Info
Title
Perfect histogram matching PCA for face recognition
Abstract
We present an enhanced principal component analysis (PCA) algorithm for improving rate of face recognition. The proposed pre-processing method, termed as perfect histogram matching, modifies the image histogram to match a Gaussian shaped tonal distribution in the face images such that spatially the entire set of face images presents similar facial gray-level intensities while the face content in the frequency domain remains mostly unaltered. Computationally inexpensive, the perfect histogram matching algorithm proves to yield superior results when applied as a pre-processing module prior to the conventional PCA algorithm for face recognition. Experimental results are presented to demonstrate effectiveness of the technique.
Year
DOI
Venue
2010
10.1007/s11045-009-0099-y
Multidim. Syst. Sign. Process.
Keywords
Field
DocType
Principal component analysis,Histogram matching,Face recognition
Frequency domain,Computer vision,Facial recognition system,Pattern recognition,Histogram matching,Adaptive histogram equalization,Artificial intelligence,Balanced histogram thresholding,Histogram equalization,Image histogram,Principal component analysis,Mathematics
Journal
Volume
Issue
ISSN
21
3
0923-6082
Citations 
PageRank 
References 
0
0.34
28
Authors
2
Name
Order
Citations
PageRank
Ana-Maria Sevcenco1133.24
Wu-Sheng Lu229624.90