Text-Independent Source Identification of Printed Documents using Texture Features and CNN Model

DOI

10.2991/978-94-6463-196-8_20How to use a DOI?

Keywords

Printers; Forgery document; LBP; K-NN; SVM; CNN

Abstract

Artificial Intelligence (AI) technologies have been used in digital forensic science to resolve disputed documents where one or more human experts would normally be contacted. The purpose of intelligent systems based on printer identification is determined printer created a specific document. Most solutions based on a text-dependent approach may be insufficient in certain scenarios. No study on text-independent based on various word images printed from various laser printer models has been done, as far as the researchers are knowledgeable. As a result, we classify the laser printer models based on the various gray scale word images. 40000-word images of four laser printer models are included in the collection. To classify the different laser printer models, the LBP (Local Binary Pattern) with KNN (K-Nearest Neighbors) and the cubic SVM (Support Vector Machine) classifiers are employed. The deep learning CNN (Convolution Neural Network) model is also used to determine the laser printer models. The experimental results of textural features and the CNN architecture are compared to recent work from a literature survey. We obtained high accuracy from K-NN and cubic SVM classifiers of 97.2% and 97.9%, respectively, and 94.3% accuracy in the CNN model.

Copyright

© 2023 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Pushpalata Gonasagi
AU  - Shivanand S. Rumma
AU  - Mallikarjun Hangarge
PY  - 2023
DA  - 2023/08/10
TI  - Text-Independent Source Identification of Printed Documents using Texture Features and CNN Model
BT  - Proceedings of the First International Conference on Advances in Computer Vision and Artificial Intelligence Technologies (ACVAIT 2022)
PB  - Atlantis Press
SP  - 250
EP  - 261
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-196-8_20
DO  - 10.2991/978-94-6463-196-8_20
ID  - Gonasagi2023
ER  -