Offline Signature Verification Based on Partial Sum of Second-Order Taylor Series Expansion

This paper presents a novel feature extraction technique which is based on partial sum of second-order Taylor Series Expansion (TSE) for offline signature verification. Partial sum of TSE is calculated with finite number of terms within a small neighborhood of a point, yields approximation for the regular function. This essentially an effective mechanism to extract the localized structural features from signature. We propose kernel structures by incorporating the Sobel operators to compute the higher order derivatives of TSE. Support Vector Machine (SVM) classifier is employed for the signature verification. The outcome of experiments on standard signature datasets demonstrates the accuracy of the proposed approach. We performed a comparative analysis for our approach with some of the popular other approaches to exhibit the classification accuracy.

[1]  Indrajit Bhattacharya,et al.  Offline Signature Verification Using Pixel Matching Technique , 2013 .

[2]  Youcef Chibani,et al.  The effective use of the one-class SVM classifier for handwritten signature verification based on writer-independent parameters , 2015, Pattern Recognit..

[3]  B. H. Shekar,et al.  Local Morphological Pattern Spectrum Based Approach for Off-line Signature Verification , 2013, PReMI.

[4]  Volker Krüger,et al.  Iris recognition by fusing different representations of multi-scale Taylor expansion , 2011, Comput. Vis. Image Underst..

[5]  Paigwar Shikha,et al.  Neural Network Based Offline Signature Recognition and Verification System , 2013 .

[6]  Babak Nadjar Araabi,et al.  Deep Multitask Metric Learning for Offline Signature Verification , 2016, Pattern Recognit. Lett..

[7]  Sargur N. Srihari,et al.  Offline Signature Verification And Identification Using Distance Statistics , 2004, Int. J. Pattern Recognit. Artif. Intell..

[8]  C. Kruthi,et al.  Offline Signature Verification Using Support Vector Machine , 2014, 2014 Fifth International Conference on Signal and Image Processing.

[9]  Algirdas Bastys,et al.  Iris recognition by local extremum points of multiscale Taylor expansion , 2009, Pattern Recognit..

[10]  Bhabatosh Chanda,et al.  Writer-independent off-line signature verification using surroundedness feature , 2012, Pattern Recognit. Lett..

[11]  Chih-Jen Lin,et al.  A Practical Guide to Support Vector Classication , 2008 .

[12]  B. H. Shekar,et al.  BLOCKWISE BINARY PATTERN: A ROBUST AND AN EFFICIENT APPROACH FOROFFLINE SIGNATURE VERIFICATION , 2017 .

[13]  Bhabatosh Chanda,et al.  A writer-independent off-line signature verification system based on signature morphology , 2010, IITM '10.

[14]  N. B. Puhan,et al.  Inter-point envelope based distance moments for offline signature verification , 2014, 2014 International Conference on Signal Processing and Communications (SPCOM).

[15]  Siyuan Chen,et al.  Machine Learning for Signature Verification , 2006, ICVGIP.

[16]  B. H. Shekar,et al.  Eigen-signature: A robust and an efficient offline signature verification algorithm , 2011, 2011 International Conference on Recent Trends in Information Technology (ICRTIT).

[17]  B. H. Shekar,et al.  Iris recognition using partial sum of second order Taylor series expansion , 2016, ICVGIP '16.

[18]  Mustafa Berkay Yilmaz,et al.  Offline signature verification using classifier combination of HOG and LBP features , 2011, 2011 International Joint Conference on Biometrics (IJCB).