2022
DOI: 10.1007/978-981-16-7136-4_19
|View full text |Cite
|
Sign up to set email alerts
|

Handwritten Signature Verification Using Transfer Learning and Data Augmentation

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
2
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
2
2
1

Relationship

0
5

Authors

Journals

citations
Cited by 5 publications
(2 citation statements)
references
References 10 publications
0
2
0
Order By: Relevance
“…The task of preprocessing is to remove invalid information that is not discriminative during the identification process and to reduce the intra-class variability between images of genuine and fake signatures [15]. In practical applications, in the offline handwritten signature system, the signer signs on paper, and then all signature data is scanned into a static image, so the signature image may have changes in the background, pen thickness, scale, rotation, etc., To improve the validity of the model, the signature image model needs to be cropped and resized.…”
Section: Preprocessingmentioning
confidence: 99%
“…The task of preprocessing is to remove invalid information that is not discriminative during the identification process and to reduce the intra-class variability between images of genuine and fake signatures [15]. In practical applications, in the offline handwritten signature system, the signer signs on paper, and then all signature data is scanned into a static image, so the signature image may have changes in the background, pen thickness, scale, rotation, etc., To improve the validity of the model, the signature image model needs to be cropped and resized.…”
Section: Preprocessingmentioning
confidence: 99%
“…Hafemann et al used deep Convolutional Neural Networks for feature learning for offline handwritten signature authentication [9]; Yapici et al used Deep Learning methods for data augmentation [10]; Tsourounis et al al-so used deep feature learning method for offline handwritten signature authentication [11]. In terms of transfer learning, Souza et al proposed a signature authentication method based on the combination of transfer learning and feature selection [12]; Tuncer et al also use transfer learning to generate deep features and perform feature selection through iterative minimum redundancy maximum correlation method before signature authentication [13]; Gupta et al combined transfer learning and data augmentation methods for offline handwritten signature authentication [14]. These methods have achieved the best authentication effect at that time.…”
Section: Introductionmentioning
confidence: 99%