Every person has a unique signature used mainly for personal identification and verification of essential documents or legal transactions. There are two kinds of signature verification: static and dynamic. Static(offline) verification is verifying an electronic or paper signature after it has been made. In contrast, dynamic(online) verification occurs as a person creates their signature on a digital tablet or a similar device. In today’s world, offline signatures play an important role. Offline signature verification and forgery detection is a challenging field with many critical issues. Signature forgery drives cooperates and business organizations to substantial financial loss and affects their security reputation. Fraud can be mainly seen in Banks, which involve important documents, legal paper works, and Government policies (LIC) that could pose a threat and be viable to forgery and its implications. So, there is a need for a system that can distinguish between genuine and forged signatures to avoid the chances of theft or fraud. Hence, we aim to create an intelligent model that gets trained with the real signature data sets and can detect a forged signature based on the available forged signature data sets using the Support Vector Machine (SVM) and K – Means algorithm. The proposed system achieves an accuracy of 95.83% for forgery detection.