Metadata and the description of digital images - PowerPoint PPT Presentation

About This Presentation
Title:

Metadata and the description of digital images

Description:

Presentation given with Roddy MacLeod (Heirot Watt University) at the conference 'NetLab and Friends,' Lund, Sweden, 10-12 April 2002. – PowerPoint PPT presentation

Number of Views:156
Avg rating:3.0/5.0
Slides: 13
Provided by: michae748
Category:

less

Transcript and Presenter's Notes

Title: Metadata and the description of digital images


1
Metadata and the description of digital images
  • Michael Day
  • UKOLN, University of Bath
  • m.day_at_ukoln.ac.uk
  • International Digital Image Symposium
  • London, 20-21 June 2002

2
Presentation outline
  • Definitions
  • The FILTER project context
  • The FILTER report, Metadata for images
  • Some conclusions

3
What is metadata?
  • Metadata is
  • Data about data
  • About both digital and non-digital objects
  • Primarily about function, e.g.
  • Resource discovery, retrieval and use
  • Data Management
  • Preservation and continuing access
  • etc.
  • Categorised as
  • Descriptive
  • Administrative
  • Structural

4
The FILTER project
  • The FILTER project is
  • Focusing Images for Learning and Teaching - an
    Enriched Resource
  • Funded as part of the JISC Learning and Teaching
    Programme
  • Partners include
  • Institute for Learning and Research Technology
    (ILRT), University of Bristol
  • Technical Advisory Service for Images (TASI)
  • Learning and Teaching Support Network (LTSN)
    Subject Centres

5
FILTER metadata review
  • Looked at selected metadata initiatives relating
    to images
  • General (e.g. DCMES, MPEG-7)
  • Cultural heritage
  • Digitisation
  • Geospatial
  • Educational
  • http//www.ukoln.ac.uk/metadata/filter/report/

6
Cultural heritage initiatives
  • Initiatives include
  • AMICO Data Dictionary
  • SCRAN Data Standard
  • SPECTRUM - the UK Museum Documentation Standard
  • VRA Core Categories
  • Specific focus on
  • discovery rights management provenance
  • Main characteristics of domain
  • not only concerned with images linked with
    digitisation initiatives describes objects as
    well as the images derived from them some
    convergence, e.g. with DCMES

7
Digitisation initiatives
  • Initiatives include
  • NISO Z39.87-2002 AIIM 20-2002 Data Dictionary --
    Technical Metadata for Digital Still Images
  • Metadata Encoding and Transmission Standard
    (METS)
  • Specific focus on
  • structure digitisation processes rights
    management digital preservation object exchange
  • Main characteristics of domain
  • fast developing area test implementations in XML

8
Geospatial initiatives
  • Initiatives include
  • FGDC Content Standard for Digital Geospatial
    Metadata (CSGDM)
  • ISO 19115 Geographical information -- Metadata
  • Specific focus on
  • geographical information spatial data GIS
  • Main characteristics of domain
  • large schemas not only concerned with images
    convergence of existing standards development on
    ISO 19115

9
Educational initiatives
  • Not specifically concerned with images
  • Initiatives include
  • IEEE LTSC Draft Standard for Learning Object
    Metadata (LOM)
  • IMS Meta-data Specification
  • DCMI Education Working Group
  • Specific focus on
  • granularity interactivity pedagogic purpose
  • Main characteristics of domain
  • format diversity links with DCMI attempts at
    achieving consensus, e.g. the Metadata for
    Education Group (MEG) in UK

10
Some conclusions (1)
  • Format diversity
  • A healthy sign?
  • metadata schemas need to be fit for purpose
  • Images are highly heterogeneous
  • but ... leads to an increased need for the
    sharing of experiences (and schemas) across
    domains -- metadata registries?
  • Some convergence within domains
  • Less convergence between domains
  • role of DCMI?

11
Some conclusions (2)
  • Interaction with CBIR
  • need for closer links between concept-based
    retrieval (e.g., that based on metadata) and
    content-based image retrieval (CBIR)?
  • Image subject matter and meaning
  • different concepts of what an image represents
  • there are some image-specific subject schemes
    (e.g., AAT, TGM, ICONCLASS) but these do not
    cover all types of image

12
Acknowledgements
  • UKOLN is funded by Resource the Council for
    Museums, Archives and Libraries, the Joint
    Information Systems Committee (JISC) of the UK
    higher and further education funding councils, as
    well as by project funding from the JISC and the
    European Union. UKOLN also receives support from
    the University of Bath, where it is based.

http//www.ukoln.ac.uk/
Write a Comment
User Comments (0)
About PowerShow.com