A white-box analysis on the writer-independent dichotomy transformation applied to offline handwritten signature verification

Abstract

High number of writers, small number of training samples per writer with high intra-class variability and heavily imbalanced class distributions are among the challenges and difficulties of the offline Handwritten Signature Verification (HSV) problem. A good alternative to tackle these issues is to use a writerindependent (WI) framework. In WI systems, a single model is trained to perform signature verification for all writers from a dissimilarity space generated by the dichotomy transformation. Among the advantages of this framework is its scalability to deal with some of these challenges and its ease in managing new writers, and hence of being used in a transfer learning context. In this work, we present a white-box analysis of this approach highlighting how it handles the challenges, the dynamic selection of references through fusion function, and its application for transfer learning. All the analyses are carried out at the instance level using the instance hardness (IH) measure. The experimental results show that, using the IH analysis, we were able to characterize "good" and "bad" quality skilled forgeries as well as the frontier region between positive and negative samples. This enables futures investigations on methods for improving discrimination between genuine signatures and skilled forgeries by considering * Corresponding author

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

“…However, it presents similar results to Souza et al. [17] and is worse when compared to the model proposed by Zois et al [27].…”

“…Table IV contains the comparison of the presented models with the state of the art methods for the GPDS-300 dataset. Souza et al [17] represents the WI-SVM trained in the original feature space.…”

“…In this paper, the SVM is used as writer-independent classifier with the following settings: RBF kernel, γ = 2 −11 and C = 1.0. The signed distance of the samples to the classifier's hyperplane is used as classifiers output [17]. Signatures were randomly selected, and a different SVM was trained for each replication (five replications were performed for each experimental configuration).…”

Interpol questioned documents review 2019–2022

“…While the research into various aspects of automated systems for various tasks related to handwriting and signature analysis is significant, one relevant textbook [ 190 ] and104 articles [ [191] , [192] , [193] , [194] , [195] , [196] , [197] , [198] , [199] , [200] , [201] , [202] , [203] , [204] , [205] , [206] , [207] , [208] , [209] , [210] , [211] , [212] , [213] , [214] , [215] , [216] , [217] , [218] , [219] , [220] , [221] , [222] , [223] , [224] , [225] , [226] , [227] , [228] , [229] , [230] , [231] , [232] , [233] , [234] , [235] , [236] , [237] , [238] , [239] , [240] , [241] , [242] , [243] , [244] , [245] , [246] , [247] , [248] , [249] , [250] , [251] , [252] , ...…”

Intrapersonal Parameter Optimization for Offline Handwritten Signature Augmentation

“…It is important to highlight that the feature descriptor must be sufficiently discriminant to measure the dissimilarity between the signatures and synthetic samples [62]. If the feature descriptor is poor, then the distinction between them will be poor as well [30].…”