Sage Journals: Discover world-class research

Abstract

Signature authentication refers to the verification whether the signature is legitimate or forged and is executed by the person who is claiming to have signed it. The main objective of the present research was to customize a deep learning-based convolutional neural network (CNN) model trained on a primary dataset for signature authentication. The model was trained, validated, and tested on the dataset of the 1400 signature images (700 genuine and 700 forged) primarily obtained from the study participants. The signature samples were equally divided into a train (1000 samples comprising 500 forged and 500 genuine signatures) and a test dataset (400 samples comprising 200 forged and 200 genuine signatures) as per the requirements of the model's training and testing procedure. The model's architecture was optimized with different hyperparameters to achieve a higher accuracy rate. The results show that the proposed model attains the training, validation and testing accuracy of 97.32%, 97.92%, and 84.5% respectively. Furthermore, other evaluation matrices were also computed, including precision, recall (sensitivity), F1-score, and specificity with the values of 85%, 84%, 84%, and 90%, respectively. The accuracy matrices provide better performance over the other existing methods. This customized CNN architecture can be trained on larger datasets and directly deployed in the field of forensic science for signature examination. The study has wide-ranging applications in the banking sector, forensic document examination, courtrooms, and beyond.

Keywords

Signature biometrics handwritten signatures deep learning convolutional neural network forgery detection signature authentication

Get full access to this article

View all access options for this article.

References

Zapata-Zapata

Arias-Londoño

Vargas-Bonilla

, et al. On-line signature verification using Gaussian mixture models and small-sample learning strategies. Rev Fac Ing Univ Antioq (En Línea) 2016; 79: 86–97.

Malallah

. Handwritten signature forgery detection using PCA and Boruta feature selection. In: 2022, 4th International Conference on Advanced Science and Engineering (ICOASE), 2022, pp.160–165. https://doi.org/10.1109/ICOASE56293.2022.10075606

Liu

Silverman

. A practical guide to biometric security technology. IT Prof 2001; 3: 27–32.

Impedovo

Pirlo

. Verification of handwritten signatures: an overview. In: 14th International Conference on Image Analysis and Processing (ICIAP 2007), 200, pp.191–196: IEEE. https://doi.org/10.1109/ICIAP.2007.4362778

Harrison

. Suspect documents their scientific examination. Reprint of 1958 ed. Chicago, USA: Nelson Wall, 1981.

Meena

Krishan

Ghosh

, et al. Is it possible to estimate sex from signatures and handwriting? A review of literature, observations, and future perspectives. Sci Nat 2023; 110: 32.

Liwicki

Schlapbach

Loretan

, et al. Automatic detection of gender and handedness from on-line handwriting. In: Proc. 13th Conf. of the Graphonomics Society, 2007, pp.179–183.

Burges

. A tutorial on support vector machines for pattern recognition. Data Min Knowl Discov 1998; 2: 121–167.

Åström

Thorell

. Graphometric variables of one’s signature: II. Correlations with measures of perception, cognitive function, and personality. Percept Mot Skills 2008; 107: 197–221.

10.

Vargas

Ferrer

Travieso

, et al. Off-line signature verification based on grey level information using texture features. Pattern Recognit 2011; 44: 375–385.

11.

Yapıcı

Tekerek

Topaloğlu

. Deep learning-based data augmentation method and signature verification system for offline handwritten signature. Pattern Anal Appl 2021; 24: 165–179.

12.

Tyagi

Agarwal

Jaiswal

, et al. Forged document detection and writer identification through unsupervised deep learning approach. Multi Tools Appl 2024; 83: 18459–18478.

13.

Shorten

Khoshgoftaar

. A survey on image data augmentation for deep learning. J Big Data 2019; 6: 1–48.

14.

Umer

Ashraf

Ullah

, et al. COVINet: a convolutional neural network approach for predicting COVID-19 from chest X- ray images. J Ambient Intell Humaniz Comput 2022; 13: 535–547.

15.

Rectified Linear Units (ReLu) in Deep Learning https://www.kaggle.com/code/dansbecker/rectified-linear-units-relu-in-deep- learning (accessed 23 April 2024).

16.

Demir

Abdullah

Sengur

, et al. A new deep CNN model for environmental sound classification. IEEE Access 2020; 8: 66529–66537.

17.

Bird

Found

Ballantyne

, et al. Forensic handwriting examiners’ opinions on the process of production of disguised and simulated signatures. Forensic Sci Int 2010; 195: 103–107.

18.

Maza-Merchán

Cordero

. Handwritten signature verification system using convolutional neural network for real-time applications. In: International Conference on Applied Technologies, 2022, pp.103–117. https://doi.org/10.1007/978-3-031-24985-3_8

19.

Lopes

Baptista

Lavado

, et al. Offline handwritten signature verification using deep neural networks. Energies 2022; 15: 7611.

20.

Tarek

Atia

. Forensic handwritten signature identification using deep learning. In: 2022 IEEE 9th International Conference on Sciences of Electronics, Technologies of Information and Telecommunications (SETIT), 2022, pp.185–190: IEEE. https://doi.org/10.1109/SETIT54465.2022.9875697

21.

Pramanik

Bag

. Handwritten Bangla city name word recognition using CNN- based transfer learning and FCN. Neural Comput 2021; 33: 9329–9341.

22.

Sharma

Bains

Verma

, et al. Novel use of logistic regression and likelihood ratios for the estimation of gender of the writer from a database of handwriting features. Aust J Forensic Sci 2023; 55: 89–106.

A customized CNN model for signature authentication—Forensic implications

Abstract

Keywords

Get full access to this article

References