Filigranes pour tous Watermarks For All

Slides:



Advertisements
Similar presentations
Patient information extraction in digitized X-ray imagery Hsien-Huang P. Wu Department of Electrical Engineering, National Yunlin University of Science.
Advertisements

Collection-level description & the Information Landscape: users evaluate strategies for resource discovery Collection Description Focus Workshop 5 Cambridge,
European Workshop on Grid-based Virtual Organisations & collaborative e-Enterprise applications Toan NGUYEN May 30th, 2003 London (UK) Business models,
Providing collections, tools and services for digital humanities A national library perspective Clément Oury Head of Digital Legal Deposit Bibliothèque.
About «Cross Border E-archive» Conference «Digital archives and historical cross border heritage» 19 June 2014, Riga, Latvia.
Mixing web and digitized archives The future of digital heritage of the World War I Valérie Beaudouin (Telecom ParisTech), Philippe Chevallier (BnF), Lionel.
Internet Vision - Lecture 3 Tamara Berg Sept 10. New Lecture Time Mondays 10:00am-12:30pm in 2311 Monday (9/15) we will have a general Computer Vision.
Maximizing Strength of Digital Watermarks Using Neural Network Presented by Bin-Cheng Tzeng 5/ Kenneth J.Davis; Kayvan Najarian International Conference.
Handwritten Character Recognition Using Artificial Neural Networks Shimie Atkins & Daniel Marco Supervisor: Johanan Erez Technion - Israel Institute of.
Multimedia for the Web: Creating Digital Excitement Multimedia Element -- Graphics.
Constructing the Memories Creating a Digital Collection Linda J. White, Digital Project Coordinator.
Robust Tools for Archiving and Preserving Digital Data Joseph JaJa, Mike Smorul, and Mike McGann Institute for Advanced Computer Studies Department of.
Associative Learning in Hierarchical Self Organizing Learning Arrays Janusz A. Starzyk, Zhen Zhu, and Yue Li School of Electrical Engineering and Computer.
OCLC Online Computer Library Center Two Paths to Interoperable Metadata Jean Godby, Devon Smith, Eric Childress DC-2003 September 29, 2003.
Case Studies Dr Lee Nung Kion Faculty of Cognitive Sciences and Human Development UNIVERSITI MALAYSIA SARAWAK.
Teaching and Learning with Technology  Allyn and Bacon 2002 Administrative Software Chapter 5 Teaching and Learning with Technology.
West Virginia University
European Organization for Nuclear Research Organisation Européenne pour la Recherche Nucléaire CDS Invenio CERN’s open source digital library information.
Chapter 6 Publishing to the iPad. Installing Software for Working with the iPad When you create layout in InDesign, you can use the Adobe Content Viewer.
Challenges for Academic Libraries in the Networked World Christine L. Borgman Professor & Presidential Chair in Information Studies UCLA & Visiting Professor.
EContentplus BERNSTEIN – THE MEMORY OF PAPERS Collaborative systems for paper expertise and history (targeted project) max. EU funding: 1,6 Mill EURO project.
The University of Florida Digital Collections Focus on Architecture Laurie N. Taylor
UKOLN is supported by: Introduction to Collections and Collection-Level Description Bridget Robinson Collection Description Focus A centre of expertise.
CBSOR,Indian Statistical Institute 30th March 07, ISI,Kokata 1 Digital Repository support for Consortium Dr. Devika P. Madalli Documentation Research &
The iPlant Collaborative Community Cyberinfrastructure for Life Science Tools and Services Workshop Discovery Environment Overview.
Dr. Claudia Fabian 27th June 2013 Piloting a National Programme for the Digitisation of Medieval Manuscripts in Germany y.
CONTENT DISCOVERY, SERVICES, AND SUSTAINED ACCESS Timothy Cole, William Mischo, Beth Sandore, Sarah Shreeves ~ University of Illinois Library
Implementing PTFS ArchivalWare at York St John University: a project under the JISC Repositories Start-up and Enhancement (SUE) strand Helen Westmancoat.
ALA Institutional Repository Update ALA Archives at the University of Illinois Urbana-Champaign Chris Prom Cara Bertram Denise Rayman.
The iPlant Collaborative Community Cyberinfrastructure for Life Science Tools and Services Workshop Discovery Environment Overview.
From small beginnings: Developing collection level description Mapping the Information Landscape Showcase day British Library Conference Centre, London,25.
Direction de l’Information Scientifique 1 Scientific and Technical Information at CNRS Laurent Romary Directeur de l’information scientifique - CNRS.
A Training Program for Shareable Metadata Metadata for You & Me is a collaboration between the University of Illinois Library and Indiana University. This.
Analysis of Classification Algorithms In Handwritten Digit Recognition Logan Helms Jon Daniele.
Creating Multimedia Repositories: new media, new metadata, new interactions…. Edinburgh Repositories Fringe, 31 Jul-01 Aug, 2008.
Tips for Training Neural Network
Research Services at La Merced Campus Objectives: Provide a framework to support researchers to develop their teaching and research. Provide advanced students.
Using Creating a Glog. GLOGSTER.COM---INTERACTIVE POSTERS GLOG- a glog is in interactive poster with elements that can be added.
Distributed Pattern Recognition System, Web-based by Nadeem Ahmed.
Standards for representing meeting metadata and annotations in meeting databases Standards for representing meeting metadata and annotations in meeting.
CLASS Metadata and Remote Sensing Extensions CLASS Data Provider’s Conference September 2005 Anna Milan, Ted.Habermann,
CERN Document Server 19 tth January 2006 CERN Document Server Jean-Yves Le Meur 19 th January 2006.
Google Books Settlement Hearing September 7, 2009, Brussels Panel 3: The Book Rights Registry Towards an International Registry Bernard Lang INRIA – AFUL.
Design and Use of Earth Observation Image Content Tools Mihai Datcu(1, 2), Daniele Cerra(1), Houda Chaabouni-Chouayakh(1), Amaia de Miguel(1), Daniela.
Digital Video Library - Jacky Ma.
Presented by Mathieu Delalandre CESR Meeting CESR, Tours, France
YugNIRO Digitization Proposal 2012
Article Review Todd Hricik.
4th Int. Conference on Watermarks in Digital Collections
Recall The Team Skills Analyzing the Problem
Chapter 12 Object Recognition
AND THE WATERMARKS OF LYON
Chapter 4 Application Software
An Information System about Research Units
Head, IT Systems Section
State-of-the-art face recognition systems
Recurrent Neural Networks
Metadata to fit your needs... How much is too much?
9.a Report on IPC-related IT systems IPC Committee of Experts 50
Similarity based on Shape and Appearance
network of simple neuron-like computing elements
Stewart Bodner OCLC Members Council May 25, 2004
Surreal Digital vs. Drawing
Office Edition Overview (Dec. 2018).
IdRef – Service of reference frames for Higher Education and Research
How Digital Humanities adds to PhD Projects
Face Recognition: A Convolutional Neural Network Approach
The Image The pixels in the image The mask The resulting image 255 X
Department of Computer Science Ben-Gurion University of the Negev
Deep Learning with Botanical Specimen Images
Presentation transcript:

Filigranes pour tous Watermarks For All A new project based on deep-learning technology and crowd-sourcing Marc H. Smith École nationale des chartes / Centre-Jean Mabillon Paris Sciences & Lettres Watermarks in digital collections 4th International Conference Vienna, 19-20 October 2017

« Science des données, données de la science » IRIS – Initiative de recherche interdisciplinaire et stratégique École nationale des chartes Christine Bénévent, Olivier Poncet, Marc Smith École des Ponts ParisTech Mathieu Aubry INRIA – Institut national de recherche en informatique et en automatique Joseph Sivic IRHT – Institut de recherche et d’histoire des textes François Bougard, Bruno Bon

Repertories of watermarks: evolution and limitations From drawings to photographs From paper to digital From single/national corpora to portals and interoperability Limitations: – Identifying watermarks: image > word > image – Number of reference images: more often “similar” than identical – Closed data, from producer to user

Filigranes pour tous Identification : image to image Deep-learning technology for image comparison Initial corpus: French watermarks > international collaboration? User interaction: image matching and database augmentation > Multiple images of (identical or variant) watermarks

Test corpus Set of homogeneous watermarks from French archives Notarial records from the Archives nationales (1650) 4 different watermarks × 61 photographs using 3 lightsheets and 3 smartphones Minimal guidelines for framing. Pages with and without writing

Test sample: four watermarks

Random sample of multiple occurrences of a watermark

Image capture and pre-processing

Image capture and pre-processing 1/6 1/6 1/6 1/6

Image capture and pre-processing 1/6 1/6 300 x 300 pixels 1/6 1/6

Deep learning Convolutional neural network: Iteration of simple operations with multiple parameters Parameters are optimized on training data, producing a different result for each watermark … Image Layer 1 Layer 2 classifier

Elementary operation of a single ‘neuron’ x = input, w = parameters

Image matching: first results Training set: 200 images (50 / watermark ) 100% correct matching Control set: 44 images (11 / watermark) 95 % correct matching (42/44) Caution: “black box” syndrome: is the matching actually based on watermarks?

Further development: the app Tools for image capture: Ruler & framing mask > scale Real-time uploading and image comparison User-uploaded images and metadata added to the database

Open questions Expanding the data set: how will the software adapt? Minimum training data set? (a single image?) Fragmentary/partially visible watermarks (sub-folio quires) Capture: close-ups vs full pages — at 300 × 300 pix ! Comparing photographs and drawings? Stimulation of crowdsourcing

Research questions Quantitative measurements Watermark variants and evolution: copies, deterioration, etc. Paper history: from production to circulation and consumption Functional distribution of formats and quality: books vs documents vs art…

marc.smith@enc-sorbonne.fr