Bee Swarm based Feature Selection for Fake and Real Fingerprint Classification using Neural Network Classifiers

摘要

With the emergent exercise of biometric authentication systems, fake and real fingerprint classification has become an attractive research area in the last decade. A number of research works have been carried out to classify fake and real fingerprints. But, most of the existing techniques did not utilize swarm intelligence techniques in their fingerprint classification system. Swarm intelligence has been widely used in various applications due to its robustness and potential in solving a complex optimization problem. The main aim of this paper is to develop a new and efficient fingerprint classification approach based on swarm intelligence with fuzzy based neural network techniques to overcome the limitations of the these classification approaches. The proposed classification methodology comprises of four steps, image preprocessing, feature extraction, feature selection and classification. This work uses efficient min-max normalization and median filtering for preprocessing, and multiple static features are extracted from Gabor filtering. Then, from the multiple static features obtained from 2D Gabor filtering, best features are selected using Artificial Bee Colony (ABC) optimization based on its searching capability. This optimization based feature selection selects only the optimal set of features which is used for classification. This would lessen the complexity and the time taken by the classifier. This approach uses Fuzzy Feed Forward Neural Network (FFFNN) for classification and its performance is compared with the SVM classifier. The performance and evaluations are performed using fingerprint images collected from FVC2000 and synthetically generated database using SFinGE. It shows that proposed work provides better results in terms of sensitivity, precision, specificity and classification accuracy.