A system for segmenting and extracting paper-based watermark designs

Hiary, Hazem; Ng, Kia

doi:10.1007/s00799-007-0008-7

Cited by 13 publications

(8 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Scientists saw a real chance of separating the watermark from the overlaying writing by using these new techniques (Zamperoni 1989, Hiary 2007. Here the paper under investigation was placed on a light table and the digital camera positioned over it (see Fig.…”

Section: Visible and Near-infrared Light Techniquementioning

confidence: 99%

Advantages and Disadvantages of Various Techniques for the Visualization of Watermarks

Meinlschmidt¹,

Märgner²

2009

Restaurator

View full text Add to dashboard Cite

This paper discusses the technical difficulties which may be encountered in the visualization indispensable in the localization and dating of papers and the investigation of watermarks. Modern digital techniques for visualizing obscured watermarks in historical documents are discussed. The advantages and disadvantages of the various digital methods will be presented and special attention will be devoted to the discriminating between writings and drawings and the underlying watermarks. As a new diagnostic method the use of heat radiation in the medium and long infrared spectrum is suggested: this creates an image of heat transmission in the medium due to the different densities and thickness variations in the paper. This type of recording is ecological, conservational and presents no health risks to the persons concerned. It can also be documented and communicated for any purpose with little expenditure of time and at high resolution.

show abstract

Section: Visible and Near-infrared Light Techniquementioning

confidence: 99%

Advantages and Disadvantages of Various Techniques for the Visualization of Watermarks

Meinlschmidt¹,

Märgner²

2009

Restaurator

View full text Add to dashboard Cite

show abstract

“…This cross-modality between the drawings and the photographs one would like to identify, is one of the main challenges of watermark recognition. Several studies have focused on localizing and extracting the pattern of a watermark from a photograph [35,34,59], which could potentially be helpful to match a photograph to a database of drawings [56]. However, these approaches are complex, not end-to-end and have not yet been demonstrated on a large scale.…”

Section: Computer Vision and Artmentioning

confidence: 99%

Spatially-Consistent Feature Matching and Learning for Heritage Image Analysis

et al. 2022

View full text Add to dashboard Cite

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

show abstract

“…The sequences of line distances for moldmate matching are computed with a combination of the discrete Fourier transform and Radon transform based on the assumption of straight and equidistant lines. Hiary et al [2] focused on the digitization, extraction, and graphical representation of watermarks. They used backlight scanning and image processing such as mathematical morphological operations to automatically extract and convert watermarks to graphical representations.…”

Section: Segmentation and Detection Of Chain Linesmentioning

confidence: 99%

“…In addition, a watermark can be embedded into the paper structure as a seal of quality and origin by placing bent metal wires on the sieve. Concerning laid paper, the distances between the parallel chain lines vary across the sieve, but are approximately 25-30 mm [2]. For every mold, the chain lines form a unique pattern.…”

Section: Introductionmentioning

confidence: 99%

ChainLineNet: Deep-Learning-Based Segmentation and Parameterization of Chain Lines in Historical Prints

et al. 2021

View full text Add to dashboard Cite

The paper structure of historical prints is sort of a unique fingerprint. Paper with the same origin shows similar chain line distances. As the manual measurement of chain line distances is time consuming, the automatic detection of chain lines is beneficial. We propose an end-to-end trainable deep learning method for segmentation and parameterization of chain lines in transmitted light images of German prints from the 16th Century. We trained a conditional generative adversarial network with a multitask loss for line segmentation and line parameterization. We formulated a fully differentiable pipeline for line coordinates’ estimation that consists of line segmentation, horizontal line alignment, and 2D Fourier filtering of line segments, line region proposals, and differentiable line fitting. We created a dataset of high-resolution transmitted light images of historical prints with manual line coordinate annotations. Our method shows superior qualitative and quantitative chain line detection results with high accuracy and reliability on our historical dataset in comparison to competing methods. Further, we demonstrated that our method achieves a low error of less than 0.7 mm in comparison to manually measured chain line distances.

show abstract

A system for segmenting and extracting paper-based watermark designs

Cited by 13 publications

References 6 publications

Advantages and Disadvantages of Various Techniques for the Visualization of Watermarks

Advantages and Disadvantages of Various Techniques for the Visualization of Watermarks

Spatially-Consistent Feature Matching and Learning for Heritage Image Analysis

ChainLineNet: Deep-Learning-Based Segmentation and Parameterization of Chain Lines in Historical Prints

Contact Info

Product

Resources

About