Multifeature Palmprint Recognitionusing Feature Level Fusion

Palmprint verification is an important tool for authentication of an individual and it can be of significant value in security and ecommerce applications. Palmprint identification has gained high impact over the other biometric modalities due to its reliability and high user acceptance. This paper presents a palmprint based identification approach which uses the Gabor wavelet to extract multiple features available on the palmprint, by employing a feature level fusion and classified using nearest neighbor approach. Here, we extract the features using wavelet entropy consist of contrast, correlation, energy, and homogeneity. The features are fused at feature levels. Palmprint matching is then performed by using nearest neighbor classifier. We selected 25 individuals’ left hand palm images every person is 5 and total is 125.Then we get every persons each palm images as a template (total 25).The remaining 100 are as the training samples. The experimental results achieve recognition accuracy for Gabor real part of 98.4%, FRR is 0.8% and FAR is 1.6%.And Recognition accuracy obtained for Gabor imaginary part of 97.63%, FRR is 0.8% and FAR is 2.4% on the publicly available database of Hong Kong Polytechnic University. Experimental evaluation using palmprint image databases clearly demonstrates the efficient recognition performance of the proposed algorithm compared with the conventional palmprint recognition algorithms. KeywordsFeature level fusion, FRR, FAR, Grey co-occurrence matrix, palmprint, recognition, Multifeature.