A new approach for inner-knuckle-print (IKP) recognition is proposed. The approach is based on the local binary pattern (LBP) features. In our algorithm, straight line neighbourhood is used to calculate the LBP features, so that more distinctive IKP features can be obtained. Moreover, as the LBP feature for each IKP sample, 59 binary images are extracted, and then matched by using a cross-correlation-based algorithm, which is developed to calculate the similarity between the IKP samples. The experiments on a finger image database which includes 2,000 images from 100 different individuals show the good performance of the proposed approach. © 2013 Springer-Verlag.
CITATION STYLE
Liu, M., Tian, Y., & Ma, Y. (2013). Inner-knuckle-print recognition based on improved LBP. In Lecture Notes in Electrical Engineering (Vol. 212 LNEE, pp. 623–630). https://doi.org/10.1007/978-3-642-34531-9_66
Mendeley helps you to discover research relevant for your work.