TY - CHAP
T1 - Biometrics identification and verification using projection-based face recognition system
AU - Moon, Hyeonjoon
AU - Kim, Jaihie
PY - 2004
Y1 - 2004
N2 - Over the last several years, numerous biometrics algorithms have been investigated for face, fingerprints, iris, and voice reocognition applications. For face recognition research, projection-based face recognition system form the basis of numerous algorithms and studies [3]. For biometrics identification and verification scenarios, we explicitly state the design decisions by introducing a generic modular face recognition system. We explored different implementations of preprocessing, feature extraction, and recognition module, and evaluate the different implementations using the FERET evaluation protocol. Our experiment includes changing the illumination normalization procedure, studying effects on algorithm performance of compressing images using JPEG and wavelet compression algorithms, and varying the number of eigenvectors in the representation. We perform series of experiments based on the standard FERET database and report results for identification and verification scenarios.
AB - Over the last several years, numerous biometrics algorithms have been investigated for face, fingerprints, iris, and voice reocognition applications. For face recognition research, projection-based face recognition system form the basis of numerous algorithms and studies [3]. For biometrics identification and verification scenarios, we explicitly state the design decisions by introducing a generic modular face recognition system. We explored different implementations of preprocessing, feature extraction, and recognition module, and evaluate the different implementations using the FERET evaluation protocol. Our experiment includes changing the illumination normalization procedure, studying effects on algorithm performance of compressing images using JPEG and wavelet compression algorithms, and varying the number of eigenvectors in the representation. We perform series of experiments based on the standard FERET database and report results for identification and verification scenarios.
UR - http://www.scopus.com/inward/record.url?scp=35048826780&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=35048826780&partnerID=8YFLogxK
U2 - 10.1007/978-3-540-24591-9_29
DO - 10.1007/978-3-540-24591-9_29
M3 - Chapter
AN - SCOPUS:35048826780
SN - 3540208275
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 380
EP - 394
BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
A2 - Chae, Kijoon
A2 - Yung, Moti
PB - Springer Verlag
ER -