This paper presents an off-line signature verification system that aims at verifying Arabic and Persian signatures. Arabic and Persian signatures have commonality in shapes, fine and general details. Moreover, both have unique general features that distinguish them from other signatures. The proposed system is based on Discrete Wavelet Transform (DWT) to extract common features to aid the verification step. This system consists of four steps: preprocessing, signature registration, feature extraction, and signature verification. Results show that the proposed system achieved good verification measures with low false acceptance rate (FAR) of 1.56%, and low average error rate of 6.23%, and false rejection rate (FRR) of 10.9%. The proposed system proved to be beneficial when compared to other works.
In this paper, we proposed an off-line Arabic and Persian signature verification system. The proposed technique is mainly divided into four steps: preprocessing, image registration, feature extraction, and finally signature verification. Preprocessing operations were performed to increase accuracy and decrease processing time. The system then used DWT to extract features from the signature image, and used logical operations with mathematical formulae to verify signatures. The experimental results were satisfactory and showed improvements over many recent works. Our system decreases the number of DWT levels and the number of required trainings, with a low FAR percentage of 1.56% and FRR of 10.9%. Future plans include enhancing the proposed technique to raise verification accuracy, and testing the technique with various shapes and types of signatures.
محتويات مواقع أعضاء هيئة التدريس بما فيها من نصوص وملفات وصور وأبحاث وأية مواد أخرى هي مسئولية عضو هيئة التدريس بالكامل بصفته صاحب الموقع وبما له من صلاحية مطلقة في الإضافة والحذف، وتخلي الجامعة مسئوليتها عن محتويات تلك المواقع.
جميع الحقوق محفوظة © عمادة تقنية المعلومات والتعليم عن بعد 2012م ــ 1433هـ.