Signature Verification System Using Deep Learning

Signature Verification System Using Deep Learning

Manik Bali1, Raghuveer Hosur2, Rohith Kumar V L3, Kiran Kumar4 ,Ms. Sowmya5
1Dept OF CS&E, MITE, Moodabidri, Mangalore

2 Dept OF CS&E, MITE, Moodabidri, Mangalore

   3 Dept OF CS&E, MITE, Moodabidri, Mangalore

4 Dept OF CS&E, MITE, Moodabidri, Mangalore

5Assistant Professor, Dept OF CS&E, MITE, Moodabidri, Mangalore

Abstract - The proposed system employs a convolutional neural network (CNN) architecture for signature feature extraction and classification. Furthermore, the system integrates preprocessing modules for signature image normalization, noise reduction, and feature extraction to enhance the robustness and accuracy of the verification process. Extensive experimentation and evaluation are conducted on benchmark datasets, including the widely used Tobacco 800 dataset and Kaggle dataset, to assess the performance of the proposed system in terms of accuracy, precision, recall, and score metrics. The results demonstrate the effectiveness and robustness of the deep learning-based signature verification system in accurately distinguishing between genuine and forged signatures.