O-F estimation based on curved Gabor Filter for fingerprint image enhancement

In the fingerprint recognition system the image quality has a high impact on the performance of the system. The presence of noise may affect the recognition rate considerably. In this paper we propose a novel preprocessing technique called O-F estimation in curved Gabor Filters for the fingerprint image quality enhancement. The orientation field is computed by two components, namely singular orientation and residual orientation, which are used for extracting the feature vector of the recognized fingerprint image. The proposed work is implemented using FVC2004 database which provides an acceptable improvement rate compared to the state of the art methods. Thus, False acceptance rate is reduced by 0.25% and False Rejected Rate is increased by 0.5%.