Ensemble Learning Approach for Handwritten Signature Recognition

Abstract

Handwritten signature recognition is a critical task in various security and authentication applications. This study proposes an ensemble learning approach to enhance the accuracy and reliability of detecting and classifying handwritten signatures. By integrating multiple individual models, including Convolutional Neural Networks (CNNs), VGG19, and MobileNetV3, the ensemble model leverages the strengths of each to achieve superior performance. The dataset used comprises both offline and online signatures, including genuine and forged samples from multiple writers. The ensemble model demonstrates a significant improvement over individual models, achieving an accuracy of 98.5%, precision of 97.5%, recall of 95.00%, and an F1-score of 95.00%. These results highlight the effectiveness of ensemble learning in capturing the complex variations in handwritten signatures, thus providing a robust solution for signature verification. The proposed method's superior performance underscores its potential for real-world applications in identity verification and fraud detection systems.