Quest for an advanced image standard: a report from the trenches

  • Authors:

  • Matthew Dougherty

  • Affiliations:

  • National Center for Macromolecular Imaging, Baylor College of Medicine, One Baylor Plaza, Houston, Texas

  • Venue:
  • Proceedings of the 2010 Roadmap for Digital Preservation Interoperability Framework Workshop
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

The advancement of science thrives on futuristic imagery. Such images can be abstractly n-dimensional and physically multimodal. They frequently consist of very large quantities of pixels that must be sub-sampled at multiple resolutions. Because there does not exist even one generic image standard for their long-term preservation and interoperations, researchers have no other option than to devise their own. Consequently, each scientific community maintains non-optimal and divergent pixel protocols for reasons of haste, convenience, performance and sometimes archiving, putting an unnecessary burden on future stewardship. Over the last five years, a number of distinct bio-imaging communities began discussing the need for a common pixel storage standard, which has also attracted the interest of astrophysicists, astronomers, and geographic information scientists. Such an advanced image design should build upon existing standards, particularly the Hierarchical Data Format (HDF5), the Motion Pictures Expert Group (MPEG) framework, and the Reference Model for an Open Archival Information System (OAIS). HDF5 is a de-facto standard in use in many fields of science and engineering because of its superlative features such as high computational performance, extensible multimodal multi-dimensional data designs, self-documentation, reliable open source software support, and archiving that is amenable to OAIS. The de-jure MPEG standard offers excellent methods for video compression, content description, intellectual property management, and multimedia interoperability. The integration of these venerable standards offers the best strategy to satisfy the immediate and long-term needs for an essential coin-of-the-realm for pixels. Described in this paper is an outline of issues that should be considered in an advanced scientific image standard.