Fingerprint Based Gender Classification Using Block-Based DCT
Each person’s fingerprint structure is unique and is developed for biometric authentication systems than others because fingerprints have advantages such as: feasible, differ from each other (distinct), permanent, accurate, reliable and acceptable all over the world for security and person identity. Fingerprints are considered legitimate proofs of evidence in courts of law all over the world. Frequency domain based fingerprint classification can be done using block based discrete cosine transform, which uses cosine as a basis function that gives energy based features of an image. We are taking dataset of 1000 male and 1000 female fingerprints. Knn classifier is used as a classifier which uses Euclidean distance measure for classification and classifies testing fingerprint as male or female fingerprint. This paper describes the overall process of above scheme. Some of dataset images are used for database creation (training images) and some for testing purpose (testing images). BBDCT transform will give the features of training sample images of dataset to create database of features which will be used as look up table for classification of unknown fingerprint and other fingerprints will be used for testing. Knn classifier will assign one of two groups to testing fingerprint.
Akhil Anjikar, Suchita Tarare, M. M. Goswami
To read the full article Download Full Article