Offline Signature Forgery Detection using Multi-Layer Perceptron

Numerous things require the participation of all parties, primarily throughout the process of establishing a person's identity, covering everything from medical care to bank account maintenance, aviation services, immigration, and other services. There are several ways to demonstrate one's identity, with employing a signature being the most well-liked one. As a means of identification system, the signature is utilized. By matching the signer's signature, recognition is still done manually. Thus, there is a significant need for a system that can evaluate and detect the signature's properties so that it can be utilized as an alternative to streamline the introduction of people's signatures. This paper discusses signature forgery detection using Multi-layer perceptron. The proposed network has been analyzed using various numbers of hidden layers and also with various classifiers. The proposed method is evaluated and the results show that it significantly outperforms the state-of-the-art methods.