Signature Detection, Restoration, and Verification: A Novel Chinese Document Signature Forgery Detection Benchmark

Yan, Kaihong; Zhang, Ying; Tang, Haoran; Ren, Chengkai; Zhang, Jian; Wang, Gaoang; Wang, Hongwei

doi:10.1109/cvprw56347.2022.00564

Cited by 10 publications

(3 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2. This study [2] 7. In this study, [7] Tuncer et al propose a novel method for handwritten signature verification that combines a deep feature warehouse with an iterative Minimum Redundancy Maximum Relevance (MRMR) approach.…”

Section: Literature Surveymentioning

confidence: 78%

A Novel Method of Fake Signature Detection Using Deep Learning Techniques

Yagvalya

2024

IJSREM

View full text Add to dashboard Cite

In today's digital world, making sure signatures are real is super important. Sometimes, though, people fake signatures, causing problems for money and legal stuff. This research looks into fixing this issue by finding a new way to catch fake signatures. We know that nowadays, a lot of things happen online, like signing documents and doing money stuff, and we need to make sure signatures are real. This paper introduces a fresh perspective by leveraging Convolutional Neural Networks (CNN), a deep learning technique. The objective is to enable the CNN model to autonomously learn and distinguish genuine signatures from counterfeits, offering a dynamic and adaptive solution to the evolving landscape of forgery. The research encompasses the development and application of a novel signature detection model, highlighting its adaptability and effectiveness. Through experimentation with diverse datasets and the optimization of hyperparameters, the approach aims to significantly enhance the accuracy of fake signature detection. The outcome of this research holds the promise of bolstering the security of digital transactions and contributing to the evolution of more robust signature verification systems. Keywords: Forged Signature, Convolutional Neural Networks (CNN), Digital Transactions, Signature Authentication

show abstract

Section: Literature Surveymentioning

confidence: 78%

A Novel Method of Fake Signature Detection Using Deep Learning Techniques

Yagvalya

2024

IJSREM

View full text Add to dashboard Cite

show abstract

“…Among a total of 260 users, 100 users wrote in Bengali and the other 160 ones wrote in Hindi, with 24 genuine signatures and 30 skilled forgeries per user. Yan et al [46] proposed ChiSig, a signature forgery detection benchmark that contains 10,242 samples from 102 users. In addition, there are two non-public Chinese signature datasets.…”

Section: (B) Offline Signature Datasetsmentioning

confidence: 99%

“…The offline handwriting images in the MSDS dataset are rendered on the acquisition devices while collecting online information, rather than being acquired by typically photographing or scanning (e.g. [15,46,44]). Therefore, the rendered handwriting may differ from the one written by pens in terms of tips, turning points, and thickness of the strokes, which may lead to changes in personal handwriting information.…”

Section: Limitationsmentioning

confidence: 99%

MSDS: A Large-Scale Chinese Signature and Token Digit String Dataset for Handwriting Verification

Zhang¹,

Jiang²,

Liu³

et al. 2022

Preprint

View full text Add to dashboard Cite

Although online handwriting verification has made great progress recently, the verification performances are still far behind the real usage owing to the small scale of the datasets as well as the limited biometric mediums. Therefore, this paper proposes a new handwriting verification benchmark dataset named Multimodal Signature and Digit String (MSDS), which consists of two subsets: MSDS-ChS (Chinese Signatures) and MSDS-TDS (Token Digit Strings), contributed by 402 users, with 20 genuine samples and 20 skilled forgeries per user per subset. MSDS-ChS consists of handwritten Chinese signatures, which, to the best of our knowledge, is the largest publicly available Chinese signature dataset for handwriting verification, at least eight times larger than existing online datasets. Meanwhile, MSDS-TDS consists of handwritten Token Digit Strings, i.e, the actual phone numbers of users, which have not been explored yet. Extensive experiments with different baselines are respectively conducted for MSDS-ChS and MSDS-TDS. Surprisingly, verification performances of state-of-the-art methods on MSDS-TDS are generally better than those on MSDS-ChS, which indicates that the handwritten Token Digit String could be a more effective biometric than handwritten Chinese signature. This is a promising discovery that could inspire us to explore new biometric traits. The MSDS dataset is available at https://github.com/HCIILAB/MSDS.

show abstract