1 / 14

Retrieval of the Ornaments from the Hand-Press Period: an Overview

Retrieval of the Ornaments from the Hand-Press Period: an Overview. Plan. About this work … Hand Press Period About Ornaments Digital Collection of Ornaments How DIA can help ? Content Based Image Retrieval Visual Comparison Conclusions and Perspectives. About this work ….

rollo
Download Presentation

Retrieval of the Ornaments from the Hand-Press Period: an Overview

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Retrieval of the Ornaments from the Hand-Press Period: an Overview

  2. Plan • About this work … • Hand Press Period • About Ornaments • Digital Collection of Ornaments • How DIA can help ? • Content Based Image Retrieval • Visual Comparison • Conclusions and Perspectives

  3. About this work … One-day Workshop 13th November 2007 CESR, Tours city, France Labs of Human Science Labs of Computer Science Computer Science People Etienne Baudrier Mickael Coustaty Mathieu Delalandre Nathalie Girard Nicholas Journet Dimosthenis Karatzas Jerome Landré Kamel Ait-Mohand Jean-Marc Ogier Nicolas Ragot Jean-Yves Ramel Human Science People Pierre Aquilon Sébastien Busson Silvio Corsini Marie-Luce Demonet Stephen Rawles Toshinori Uetani CESR

  4. Hand Press Period (1/2) The Hand-Press period runs from around 1454 (approximate date of Gutenberg’s invention) to through the first half of the nineteenth century (when mechanized presses started to appear). Hand Press 1454 Gutenberg half 18th mechanized presses character matrix hand press a hand-press book

  5. Hand Press Period (2/2) Mathematics, medicine, history, music, religion, literature, etc. • HPB Database • http://www.cerl.org/ • 22 European libraries • 1450 - half 19th • 3 Millions books Trinity old library (Dublin, Ireland) 16th - today

  6. About Ornaments (1/2) Ornaments in pages Categories of ornaments to start a paragraph “lettrine” trademark of a printing house “fleuron” to close a part or a chapter “cul de lampe” ornaments text to epitomize a concept , or to represent a person, such as a king or saint. “emblème”

  7. About Ornaments (2/2) Part of ornaments in books (BVH dataset, 46 books) Hand Press books are composed for a large part of ornaments. Pictures were a powerful mean of communication at this period due to the low education level of people. sciences, medical, religion …

  8. Digital Collections of Ornaments (1/2) Digitalization Pre-processing (deskew, lighting correction, filtering, cropping…) Layout analysis and segmentation [Ramel’07] Expert Classification using thesaurus icon class encoding of an emblem image

  9. Digital Collections of Ornaments (2/2) Other smallest datasets are ArtDico, Canadian heraldry, Printers' Devices, etc. • Collections of ornaments are small in regard to mass digitalization collections (e.g. Million Book Project), two main reasons: • Mass digitalization projects are thought in terms of OCR only (layout analysis aims to perform text/graphics separation, final electronic documents are “ASCII code”, no use of high-level document model) •  Digitalization programs should consider better the graphics aspects. • Classification using thesaurus by human experts is time consuming (15-20 mn per image) •  Collaborative platforms, integrating DIA components, can help in.

  10. 1555-1578 Printing house tampon 1511-1542 exchange copy 1531-1548 1497-1507 How DIA can help ? (1/2) noise Redundancy of ornaments in books Tracking of plugs Vascosan 1555 Marnef 1576 precision offset A same block used in 2 books skewing A duplicated block weak resolution, lossy compression scalability, mass of data scaling

  11. How DIA can help ? (2/2) DB1 R1 CBIR Visual Comparison R2 Digital Collections Of ornaments DB2 Meta --- R3 Meta DBn Query image Meta retrieval results submit a query comparison • Context information • Publication dates • Publication places • Practices of printers • … visualization assign previous classification Meta

  12. Content Based Image Retrieval Bigun’96 • Ideal method • High precision (weak difference) • Robust (noise, skew, offset) • Invariant to scale • Fast comparison (online, mass of data) • Scalable h w h w Radiogram 0° Radiogram 90° Expert set resolution analysis Run Length Encoding Orientation Radiograms Detection of key points (Haris) Histogram centering Hausdorff distance between images Zernike moments (local template) Fourier Descriptors SVM classification Nearest points compared with a likelihood estimation RLE Comparison Euclidean Distance Comparison Chen’03 Baudrier’08 Delalandre’07

  13. Visual Comparison Beusekom’07 • Ideal method • Highlight pertinent differences • Make an hypothesis of relative dating • Invariant to scale • Robust (noise, skew, offset) Detection of points of interest (connected components) Pixel to Pixel Difference Map (PPDMap) BlockA#1 PPDMap BlockA#2 LDMap Baudrier’07 Image Registration Visualization Method Equivalent ellipse computation (first image moments) Local Dissimilarity Map (LDMap)

  14. Conclusions and Perspectives • Large ornament material is available, but there is few digital collections • Digitalization programs should consider better the graphics aspects. • Collaborative platforms, integrating DIA components, can help in. • Two database levels (with, without thesaurus classification) • DIA components • CBIR systems (orientation signature, points of interest, image distance, compressed representation) • Lack of evaluation of the methods make difficult the comparison • To define benchmark datasets (time, precision/recall) • Methods propose a tradeoff between complexity/precision, possible combination • Visual Comparison (registration, PPDMap, LDMap) • Hard point is the registration, user interaction could help in

More Related