Academic Journals Database
Disseminating quality controlled scientific knowledge

Fingerprint Matching by Extracting GLCM Features

ADD TO MY LIST
 
Author(s): Benazir . K.K | Vijayakumar

Journal: International Journal of Computer Applications
ISSN 0975-8887

Volume: icwet;
Issue: 1;
Date: 2012;
Original page

Keywords: Fingerprint matching | GLCM | Median filtering | Euclidean distance

ABSTRACT
Fingerprint matching is an important and challenging problem in fingerprint recognition.Even though so many different methods are there, it has been learned from studies that a better feature extraction technique may leads to very good results. In this paper, we have improved the efficiency of fingerprint matching by combining GLCM based feature extraction with Euclidean based matching. Co-occurrence matrices can be used to extract features from the fingerprint image because they are composed of regular texture patterns. First, the fingerprint image is preprocessed and a unique reference point is determined to secure a Region-of-Interest (ROI). Four co-occurrence matrices are computed from the ROI with a predefined set of parameters. A feature vector consisting of 16 features are used to match the input image with different types of images stored in the database. The fingerprint matching is based on the Euclidean distance between the two corresponding fingerprints and hence is extremely fast. The validity of newly derived algorithms is tested on fingerprint images of Db1_a&Db1_bof FVC2002database. A very good result of 97% of matching is achieved.The method significantly reduces the memory cost and processing time associated with verification, primarily because of the efficient use of GLCM feature extraction. The experimental results and the ROC curves demonstrate the effectiveness of the proposed method, concerning the feature extraction of ROI, especially in low quality images.
Why do you need a reservation system?      Affiliate Program