Learning deep features for source color laser printer identification based on cascaded learning

Kim, Do-Guk; Hou, Jong-Uk; Lee, Heung-Kyu

doi:10.1016/j.neucom.2019.07.084

Cited by 15 publications

(8 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Therefore, to construct a deep learning network that can quickly adapt to various conditions so as to obtain high quality scanned inverse halftone images is urgently needed. Recently, Kim et al [46] proposed a Halftone Color Decomposing Convolutional Neural Network (HCD-CNN) that can better decompose the photographed image or scanned image as C, M, Y, and K halftone images. Inspired by the HCD-CNN, the scanned halftone image can be firstly decomposed as CMYK images, and then inversed as a continuous-tone image using an inverse halftone method for digital halftone images.…”

Section: Adaptive Inverse Halftoning For Scanned Halftone Imagesmentioning

confidence: 99%

Inverse Halftoning Methods Based on Deep Learning and Their Evaluation Metrics: A Review

Zhang

Wang

et al. 2020

Applied Sciences

View full text Add to dashboard Cite

Inverse halftoning is an ill-posed problem that refers to the problem of restoring continuous-tone images from their halftone versions. Although much progress has been achieved over the last decades, the restored images still suffer from detail loss and visual artifacts. Recent studies show that inverse halftoning methods based on deep learning are superior to other traditional methods, and thus this paper aimed to systematically review the inverse halftone methods based on deep learning, so as to provide a reference for the development of inverse halftoning. In this paper, we firstly proposed a classification method for inverse halftoning methods on the basis of the source of halftone images. Then, two types of inverse halftoning methods for digital halftone images and scanned halftone images were investigated in terms of network architecture, loss functions, and training strategies. Furthermore, we studied existing image quality evaluation including subjective and objective evaluation by experiments. The evaluation results demonstrated that methods based on multiple subnetworks and methods based on multi-stage strategies are superior to other methods. In addition, the perceptual loss and the gradient loss are helpful for improving the quality of restored images. Finally, we gave the future research directions by analyzing the shortcomings of existing inverse halftoning methods.

show abstract

Section: Adaptive Inverse Halftoning For Scanned Halftone Imagesmentioning

confidence: 99%

Inverse Halftoning Methods Based on Deep Learning and Their Evaluation Metrics: A Review

Zhang

Wang

et al. 2020

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Kondisi baik ini sekaligus menjadi tantangan, karena potensi pemalsuan semakin besar [2]. Duplikasi atau pemalsuan suatu dokumen yang menggunakan alat cetak berkualitas tinggi, hasil cetaknya sulit dibedakan secara visual antara dokumen cetak asli dengan palsu meskipun pemalsuan menggunakan printer jenis dan tipe yang berbeda [3]. Kualitas tinggi dari alat cetak akan menghasilkan dokumen cetak palsu cenderung identik dengan dokumen cetak asli, membedakan secara visual sulit dilakukan.…”

Section: Pendahuluanunclassified

“…Berdasarkan arah gradien x dan y, maka untuk Gx dengan menggunakan pendekatan diferensial horizontal terhadap x maka fungsi f(x,y) diunjukkan pada Persamaan (3).…”

Section: Gambar 2 Akuisisi Sampel Karakter Dokumen Cetak Asli (Known)...unclassified

Verifikasi Dokumen Cetak Menggunakan Metode Edge Detection-GLCM Dan K-Mean Clustering

Kurniati¹,

Huizen²

2020

dinamik

View full text Add to dashboard Cite

Dokumen cetak merupakan data digital yang tervisualisasi di kertas. Dokumen cetak mudah dibuat sekaligus dipalsukan. Cara mengetahui keasliannya dengan verifikasi ciri. Setiap dokumen cetak terdapat ciri unik berasal dari printer yang digunakan. Mengembangkan metode untuk menentukan keaslian dokumen cetak dari printer jenis dan tipe sama menjadi tantangan untuk diselesaikan pada penelitian ini. Proses verifikasi menggunakan tahapan akuisisi, ekstraksi, pengenalan pola dan hasil verifikasi. Data uji menggunakan 6 printer jenis dan tipe sama, masing-masing menggunakan 3 lembar sampel, untuk karakter dipilih 8. Hasilnya pengujian menunjukkan nilai EER (equal error rate) sebesar 0,18 dari nilai tersebut digunakan menentukan nilai ambang verifikasi yaitu 80% dari nilai kedekatan. Verifikasi dengan kedekatan diatas 80% berarti dokumen cetak dinyatakan asli (accepted) sedangkan kedekatan dibawah 80% berarti dokumen cetak dinyatakan palsu (rejected). Pengujian menggunakan 20 varian dokumen cetak diperoleh nilai kedekatan tertinggi 97,7% dan terendah 84,1%. Berdasarkan pengujian diperoleh suatu kesimpulan bahwa meskipun pemalsuan menggunakan printer dengan jenis dan tipe yang sama, model mampu memilah dan menentukan suatu dokumen cetak asli ataupun palsu.

show abstract

“…The document analysis community has recently started working towards replacing batch scanners by smartphone cameras [6]. In document forensics, a very recent approach proposed a method for source identification of colored images printed by color laser printers [7]. There are significant differences between the working of color and 'black-and-white' (grayscale) printers.…”

Section: Introductionmentioning

confidence: 99%

Source Printer Identification from Document Images Acquired using Smartphone

Joshi,

Saxena,

Khanna

2020

Preprint

View full text Add to dashboard Cite

Vast volumes of printed documents continue to be used for various important as well as trivial applications. Such applications often rely on the information provided in the form of printed text documents whose integrity verification poses a challenge due to time constraints and lack of resources. Source printer identification provides essential information about the origin and integrity of a printed document in a fast and costeffective manner. Even when fraudulent documents are identified, information about their origin can help stop future frauds. If a smartphone camera replaces scanner for the document acquisition process, document forensics would be more economical, user-friendly, and even faster in many applications where remote and distributed analysis is beneficial. Building on existing methods, we propose to learn a single CNN model from the fusion of letter images and their printer-specific noise residuals. In the absence of any publicly available dataset, we created a new dataset consisting of 2250 document images of text documents printed by eighteen printers and acquired by a smartphone camera at five acquisition settings. The proposed method achieves 98.42% document classification accuracy using images of letter 'e' under a 5×2 cross-validation approach. Further, when tested using about half a million letters of all types, it achieves 90.33% and 98.01% letter and document classification accuracies, respectively, thus highlighting the ability to learn a discriminative model without dependence on a single letter type. Also, classification accuracies are encouraging under various acquisition settings, including low illumination and change in angle between the document and camera planes.

show abstract

Learning deep features for source color laser printer identification based on cascaded learning

Cited by 15 publications

References 20 publications

Inverse Halftoning Methods Based on Deep Learning and Their Evaluation Metrics: A Review

Inverse Halftoning Methods Based on Deep Learning and Their Evaluation Metrics: A Review

Verifikasi Dokumen Cetak Menggunakan Metode Edge Detection-GLCM Dan K-Mean Clustering

Source Printer Identification from Document Images Acquired using Smartphone

Contact Info

Product

Resources

About