Classification of Fingerprint Images Using Neural Networks Technique

Ebtesam AlShemmary


Automatic fingerprint identification is one of the most important biometric technologies. In order to efficiently match fingerprints in a large database, an indexing scheme is necessary. Fingerprint classification, which refers to assigning a fingerprint image into a number of pre-specified classes, provides a feasible indexing mechanism. In the field of criminal investigation the task of classifying fingerprints consumes much time and labor. Numerous attempts have been made to automate the classification process using conventional image processing techniques but very few have been embraced by law enforcement agencies due to their limited successes in solving the problem. The reemergence of interest in neural networks in recent years has caught the attention of those involved in fingerprint recognition as they begin to recognize the potential advantages of a neural network approach. In this paper, we introduce a new approach to fingerprint classification based on both singularities and neural network analysis. Since noise exists in most of the fingerprint images including those in the NIST databases which are used by many researchers, it is difficult to get the correct number and position of the singularities such as core or delta points which are widely used in current structural classification methods. The problem is we may miss the true singular points and/or get false singular points due to the poor quality of fingerprint images. Classification based on exact pair of singularities will fail in such conditions. This paper presents some intermediate results on fingerprint classification adopting a neural network as decision stage, in order to evaluate the performance of automatic fingerprint classification using neural network techniques.


Fingerprint Image, Feature Extraction, Classification, Neural Networks

Full Text:



  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.

JOE ©: World Science Publisher United States