HPCCprojects:Lithology Image Strips Extraction for the Ocean Drilling Program

Revision as of 08:27, 14 October 2019 by WikiSysop (talk | contribs) (→‎Publications and presentations)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Lithology Image Strips Extraction for the Ocean Drilling Program

Project description

The Ocean Drilling Program drilled 653 seabed sites and recovered 236km of core in the years 1985-2003. The core recoveries were photographed, but on a frame in varying light conditions. Modern visualisation software requires image strips, which today are scanned electronically. This project extracted image strips compatible with the modern scannings, from the legacy photos.


The photos show in great detail the sediments, rock and structures that were sampled by the IODP in all oceans of the world, extending back over 120My of earth history.

The project objective was to bring this huge resource of valuable data up to modern standards and into high useability.



Models in use

The Python programs for image analysis and strip extraction were written by CJ Jenkins, INSTAAR. They are available on request.


Over 90% of the strips available were successfully extracted, in RGB color, usually at 40dpi (0.5mm) resolution. The collection - 2 TB of data - is now publicly available at servers and in core visualisation software.

  • Borehole Research Group, LDEO, Columbia University, USA,
  • ANDRILL, University of Nebraska, USA,
  • PANGAEA Database, MARUM, Bremen, DEU,
  • and their numerous database visitors.

By using the HPCC facility, the processing run time for the 36,000 photos and 1TB of input data was cut to one sixth.



Through the Consortium for Ocean Leadership, based on NSF funding to the Integrated Ocean Drilling Program.

Publications and presentations

Several public lectures and workshops have been given in the USA, Germany and Japan.


Poster: Image Strips from IODP Legacy Datasets for use in Visualization Software