Title | ||
---|---|---|
PCA-Based Face Verification and Passport Code Recognition Using Improved FKCN Algorithm |
Abstract | ||
---|---|---|
In this paper, we proposed methods to recognize passport codes and verify face for efficient management of inbound and outbound passengers and swift immigration control. A passport image can be slanted because the image is acquired by scanning. Therefore gradient correction is very important in character extraction, character recognition, and face verification processes. Gradient correction of a scanned passport image is performed based on center points and horizontal line of both side tips of extracted line of codes applying horizontal smearing method to the scanned passport image. Individual code of passport is extracted using 8-directional edge tracking algorithm and then passport codes are recognized by improved FKCN (Fuzzy Kohonen Clustering Networks) using frequency. Normalized gray-level correlation matching algorithm is used to extract face region and then feature vector of face region is extracted using PCA algorithm. Finally we can check whether face of a passport is forgery or not by comparing the extracted feature vector of face region with the face information of a passport holder extracted from passport database using recognized passport codes. From the experiment on face-forged passports and slanted passport images with our proposed methods in order to evaluate performance of the proposed passport code recognition and face verification methods, we could see the proposed methods is more efficient than conventional methods in passport code recognition and face verification. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1109/ISDA.2008.247 | ISDA (2) |
Keywords | Field | DocType |
passport code recognition,face region,proposed passport code recognition,passport image,slanted passport image,improved fkcn algorithm,scanned passport image,passport holder,face-forged passport,passport database,pca-based face verification,passport code,data mining,fuzzy set theory,pca,mathematical model,feature vector,principal component analysis,lines of code,face,feature extraction,face recognition | Normalization (statistics),Computer science,Self-organizing map,Fuzzy set,Artificial intelligence,Cluster analysis,Horizontal line test,Facial recognition system,Computer vision,Feature vector,Pattern recognition,Algorithm,Feature extraction,Machine learning | Conference |
Citations | PageRank | References |
0 | 0.34 | 4 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
kwangbaek kim | 1 | 110 | 43.94 |
Am-Suk Oh | 2 | 8 | 4.03 |
Young Woon Woo | 3 | 31 | 8.39 |