Block Induced Signature Generative Adversarial Network (BISGAN): Signature Spoofing Using GANs and Their Evaluation

TL;DR

This work focuses on creating a generator that produces forged samples that achieve a benchmark in spoofing signature verification systems and advocates generator-focused GAN architectures for spoofing data quality that aid in a better understanding of biometric data generation and evaluation.

Abstract

Deep learning is actively being used in biometrics to develop efficient identification and verification systems. Handwritten signatures are a common subset of biometric data for authentication purposes. Generative adversarial networks (GANs) learn from original and forged signatures to generate forged signatures. While most GAN techniques create a strong signature verifier, which is the discriminator, there is a need to focus more on the quality of forgeries generated by the generator model. This work focuses on creating a generator that produces forged samples that achieve a benchmark in spoofing signature verification systems. We use CycleGANs infused with Inception model-like blocks with attention heads as the generator and a variation of the SigCNN model as the base Discriminator. We train our model with a new technique that results in 80% to 100% success in signature spoofing. Additionally, we create a custom evaluation technique to act as a goodness measure of the generated forgeries. Our work advocates generator-focused GAN architectures for spoofing data quality that aid in a better understanding of biometric data generation and evaluation.

Figures (7)

• Figure 1: Overall architecture of BISGAN
• Figure 2: Generator architecture of BISGAN
• Figure 3: Discriminator of BISGAN
• Figure 4: Abstract representation of achievement of new training technique.
• Figure 5: Comparison of generated forgeries of models.