期刊名称：International Journal of Computer Trends and Technology
电子版ISSN：2231-2803
出版年度：2018
卷号：61
期号：3
DOI：10.14445/22312803/IJCTT-V61P122
出版社：Seventh Sense Research Group
摘要：The most widely recognized secure personal biometric authentication is handwritten signature. Most organizations, primarily concentrate on the visual appearance of the handwritten signature for confirmation purposes. Many archives, for ex ample, forms, contracts, bank cheques, and credit card transactions require the handwritten signature. Main challenging issues in the system is features are used for recognizing forged and genuine signatures. This article presents a technique for offline hand written signature verification by exploiting the advantage of low complexity ordereddither block truncation coding (ODBTC) for the generation of image content descriptor and Local binary patterns (LBP). LBP was widely used as a robust illumination invariant feature descriptor. The system consists of preprocessing, feature vector extraction, training and classification or verification stage. In the preprocessing stage ODBTC compresses a signature image into corresponding quantizers and bitmap image. The Bit Pattern Features (BPF) is generated from ODBTC encoded data streams and LBP feature is extracted from input image. In the training stage a set of reference, signature images for each person is used. The mean vector of the set of feature vector is used for the verification purpose. The relative distance measure was used for classification. The proposed system is executed and tested utilizing GPDS database. The performance of the system is measured and experimental result shows convenience and viability of the proposed system.
关键词：Signature Verification; Matching processes; ODBTC features LBP; BPF; and relative distance.
Loading...
