An Investigation of Feature Selection and Transfer Learning for Writer-Independent Offline Handwritten Signature Verification

TL;DR

This paper explores feature selection and transfer learning in writer-independent offline handwritten signature verification, demonstrating how validation strategies can prevent overfitting and improve model transferability across datasets.

Contribution

It introduces a global validation strategy with an external archive to control overfitting during feature selection and shows the selected features are effective for transfer learning.

Findings

Overfitting occurs without validation during feature selection.

Global validation with an external archive reduces overfitting.

Selected features are effective in transfer learning scenarios.

Abstract

SigNet is a state of the art model for feature representation used for handwritten signature verification (HSV). This representation is based on a Deep Convolutional Neural Network (DCNN) and contains 2048 dimensions. When transposed to a dissimilarity space generated by the dichotomy transformation (DT), related to the writer-independent (WI) approach, these features may include redundant information. This paper investigates the presence of overfitting when using Binary Particle Swarm Optimization (BPSO) to perform the feature selection in a wrapper mode. We proposed a method based on a global validation strategy with an external archive to control overfitting during the search for the most discriminant representation. Moreover, an investigation is also carried out to evaluate the use of the selected features in a transfer learning context. The analysis is carried out on a writer-independent approach on the CEDAR, MCYT and GPDS datasets. The experimental results showed the presence of overfitting when no validation is used during the optimization process and the improvement when the global validation strategy with an external archive is used. Also, the space generated after feature selection can be used in a transfer learning context.