Advances in Computer Science and Engineering Volume 7, Number 1, 2011, Pages 25-35 This paper is available online at http://pphmj.com/journals/acse.htm © 2011 Pushpa Publishing House : phrases and Keywords biometrics, discrete wavelet transform, support vector machine, fusion. Received January 23, 2011 AN IMPROVED PROPOSED APPROACH FOR HANDWRITTEN ARABIC SIGNATURE RECOGNITION MUZHIR SHABAN AL-ANI and MAHA MAHMOUD AL-SAIDI Department of Computer Science Amman Arab University Jordan e-mail: muzhir_shaban@yahoo.com; maha_alsaidi@yahoo.com Abstract Handwritten signature recognition plays an important part in our life and especially in security applications. The proposed handwritten signature recognition system is implemented via many steps such as Discrete Wavelet Transform (DWT), feature vector generation, fusion between feature vectors, then applying Support Vector Machine (SVM). These steps are applied for identification and verification of signatures. The results obtained from the verification process are better than the results obtained from the identification under the same circumstances. 1. Introduction Automatic handwritten signature has been studied for decades. Signature is an image carrying certain pattern of pixels that pertains to a specific individual [1]. Handwritten signature recognition is a behavioral biometric technique for personal identification [2]. Biometrics is the study of methods for uniquely recognizing humans based upon one or more intrinsic physiological or behavioral traits such as face, iris, hand, hand- veins, ear, handwriting, speech, keystroke dynamics, gait patterns, etc. Physical modalities like face, iris, hand, hand-veins, ear, etc. are more invasive and need cooperative subjects. Behavioral modalities such as handwriting (including