In recent years, the importance of biometrics in authentication and identification has emerged. Both government and private agencies make use of automated human identification using different biometric traits. Signature based authentication has been used for many years. This paper presents an automated static signature recognition approach. We have used Gabor filter for preprocessing of signature images. For matching purpose, artificial neural networks are used, which are trained by the back-propagation learning algorithm. The results show that the given approach yields a good solution to automated signature recognition problem because of its high accuracy (99.5%) and FRR (68%). On the other hand, computational time of the proposed technique is less than the previous approaches.