Single Classifier-Based Passive System for Source Printer Classification Using Local Texture Features

@article{Joshi2018SingleCP,
  title={Single Classifier-Based Passive System for Source Printer Classification Using Local Texture Features},
  author={Sharad Joshi and Nitin Khanna},
  journal={IEEE Transactions on Information Forensics and Security},
  year={2018},
  volume={13},
  pages={1603-1614}
}
  • Sharad Joshi, N. Khanna
  • Published 22 June 2017
  • Computer Science
  • IEEE Transactions on Information Forensics and Security
An important aspect of examining printed documents for potential forgeries and copyright infringement is the identification of the source printer as it can be helpful for detecting forged documents and ascertaining the leak. This paper proposes a system for classification of source printer from scanned images of printed documents using all the printed letters simultaneously. The proposed system uses local texture patterns-based features and a single classifier for classifying all the printed… Expand
Source Printer Classification Using Printer Specific Local Texture Descriptor
TLDR
A novel printer specific local texture descriptor is introduced based on small linear-shaped structures composed of pixels having similar intensity and gradient that outperforms state-of-the-art algorithms for characters printed in the same font. Expand
Source Printer Identification using Printer Specific Pooling of Letter Descriptors
The digital revolution has replaced the use of printed documents with their digital counterparts. However, many applications require the use of both due to several factors, including challenges ofExpand
Passive classification of source printer using text-line-level geometric distortion signatures from scanned images of printed documents
TLDR
A set of features for characterizing text-line-level geometric distortions is proposed and a novel system to use them for identification of the origin of a printed document and gives much higher accuracy under small training size constraint is presented. Expand
Printer source identification by feature modeling in the total variable printer space
TLDR
A new method is presented for printer source identification by modeling the primary Local Binary Pattern features in the total variable printer space and extracting the secondary features based on the joint factor analysis, which eliminates the requirement for the majority voting algorithm and reduces the computational cost of the classification process. Expand
A computational approach for printed document forensics using SURF and ORB features
TLDR
A classifier-based model to identify the source printer and classify the questioned document in one of the printer classes is proposed and can efficiently classify the questioning documents to their respective printer class. Expand
Printer Forensics Based on Identity Vectors of Image Texture Segmentation
TLDR
The effect of using the texture of different regions and changing the patch dimensions using the SVM (Support Vector Machine) classifier through simulation has been carefully investigated and results show that only by refining the basic features of LBP the authors achieved 99.05% accuracy, more than the latest research in this field. Expand
Document Forgery Detection using Printer Source Identification—A Text-Independent Approach
TLDR
A text-independent approach for effective characterization of source printer, using deep visual features using pre-trained Convolutional Neural Networks (CNNs), achieves significant recognition results on a dataset of 1200 documents from 20 different printers. Expand
Source Printer Identification from Document Images Acquired using Smartphone
TLDR
A single CNN model is proposed to learn from the fusion of letter images and their printer-specific noise residuals, in the absence of any publicly available dataset, to learn a discriminative model without dependence on a single letter type. Expand
Forensic document examination system using boosting and bagging methodologies
TLDR
This paper aims to fix the document source printer using passive approach using a combination of KPNF, speeded up robust features, and oriented FAST rotated and BRIEF with adaptive boosting and adaptive boosting methodology. Expand
Source Classification Using Document Images from Smartphones and Flatbed Scanners
TLDR
This work utilizes local tetra patterns to capture unique device-specific signatures from images of printed documents and uses all characters to train a single classifier thereby, reducing the amount of training data required. Expand
...
1
2
3
...

References

SHOWING 1-10 OF 50 REFERENCES
Passive classification of source printer using text-line-level geometric distortion signatures from scanned images of printed documents
TLDR
A set of features for characterizing text-line-level geometric distortions is proposed and a novel system to use them for identification of the origin of a printed document and gives much higher accuracy under small training size constraint is presented. Expand
Printer Identification Using Supervised Learning for Document Forgery Detection
  • Sarah Elkasrawi, F. Shafait
  • Engineering, Computer Science
  • 2014 11th IAPR International Workshop on Document Analysis Systems
  • 2014
TLDR
A method to automatically identify source printers using common-resolution scans (400 dpi) that depends on distinctive noise produced by printers, which is important in forgery detection. Expand
Document Authentication Using Printing Technique Features and Unsupervised Anomaly Detection
TLDR
A system using the difference in edge roughness to distinguish laser printed ages from inkjet printed pages is presented, and shows that the presented feature extraction method achieves the best outlier rank score in comparison to state-of-the-art features. Expand
A Novel Multi-size Block Benford's Law Scheme for Printer Identification
TLDR
A forensic technique based on the Benford's law is proposed to identify the printer's brand and model from the printed-and-scanned images at which the first digit probability distribution of multi-size block DCT coefficients are extracted that constitutes a feature vector as the input to support vector machine (SVM) classifier. Expand
Text-independent printer identification based on texture synthesis
TLDR
Experimental results demonstrate that the proposed method can achieve a high recognition rate and provide a new platform for printer identification independent on the content of documents. Expand
Identifying Color Laser Printer Using Noisy Feature and Support Vector Machine
  • Hae-Yeoun Lee, Jung-Ho Choi
  • Engineering
  • 2010 Proceedings of the 5th International Conference on Ubiquitous Information Technologies and Applications
  • 2010
Color laser printers are nowadays abused to forge official documents and bills. Identifying the source used to print documents will be a step for digital media forensics. Each printer vender appliesExpand
Application of Principal Components Analysis and Gaussian Mixture Models to Printer Identification
TLDR
Different classification algorithms to identify an unknown printer are described and intrinsic feature extraction method, based on frequency domain analysis of the one dimensional projected signal, is described. Expand
Printer identification based on graylevel co-occurrence features for security and forensic applications
TLDR
The use of image texture analysis to identify the printer used to print a document is described and a set of features that can be used to provide forensic information about a document are described. Expand
Laser printer attribution: exploring new features and beyond.
TLDR
Novel techniques for laser printer attribution are proposed that outperform techniques described in the literature and present near-perfect classification accuracy being very promising for deployment in real-world forensic investigations. Expand
Digital forensics of printed source identification for Chinese characters
TLDR
The superior testing performance demonstrates that the proposed identification method is very useful for source laser printer identification and explores the optimum feature subset by using feature selection techniques and use support vector machine (SVM) to identify the source model of the documents. Expand
...
1
2
3
4
5
...