1-D CNN-Based Online Signature Verification with Federated Learning
Lingfeng Zhang, Yuheng Guo, Yepeng Ding, Hiroyuki Sato
TL;DR
The paper tackles privacy concerns in online signature verification by introducing a privacy-preserving federated learning framework that deploys a lightweight 1-D CNN as the verifier. A central cloud coordinator orchestrates synchronous training across multiple edge agents, aggregating local updates via FederatedAveraging to form a robust global model without sharing raw signature data. Key findings show that the 1-D CNN achieves strong verification performance (central EER $3.33\%$, accuracy $96.25\%$) and that FL supports lightweight local training, effective transfer from initialization data, and scalability to multiple agents, with near-central performance observed for at least 10 agents. The work demonstrates practical impact by enabling privacy-aware, scalable biometric verification suitable for real-world deployments while maintaining competitive accuracy and low communication overhead.
Abstract
Online signature verification plays a pivotal role in security infrastructures. However, conventional online signature verification models pose significant risks to data privacy, especially during training processes. To mitigate these concerns, we propose a novel federated learning framework that leverages 1-D Convolutional Neural Networks (CNN) for online signature verification. Furthermore, our experiments demonstrate the effectiveness of our framework regarding 1-D CNN and federated learning. Particularly, the experiment results highlight that our framework 1) minimizes local computational resources; 2) enhances transfer effects with substantial initialization data; 3) presents remarkable scalability. The centralized 1-D CNN model achieves an Equal Error Rate (EER) of 3.33% and an accuracy of 96.25%. Meanwhile, configurations with 2, 5, and 10 agents yield EERs of 5.42%, 5.83%, and 5.63%, along with accuracies of 95.21%, 94.17%, and 94.06%, respectively.