JHelioviewer - Visualizing large sets of solar images using JPEG 2000

JHelioviewer – Visualizing large sets of solar images using JPEG 2000 Daniel M¨uller1, George Dimitoglou2, Benjamin Caplins2, Juan Pablo Garc´ıa Ortiz3, Benjamin Wamsler4, Keith Hughitt5, Alen Alexanderian6, Jack Ireland5, Desmond Amadigwe2, Bernhard Fleck1 1European Space Agency c/o NASA Goddard Space Flight Center, Mailcode 671.1, Greenbelt, MD 20771 2Department of Computer Science, Hood College, Frederick, MD 21701 3Computer Architecture and Electronics Department, University of Almer´ıa, Spain 4Fakult¨at f¨ur Mechatronik und Medizintechnik, Hochschule Ulm, Germany 5ADNET Systems Inc., NASA Goddard Space Flight Center, Mailcode 671.1, Greenbelt, MD 20771 6Department of Mathematics and Statistics, University of Maryland, Baltimore County, Baltimore, MD 21250 Across all disciplines that work with image data – from astrophysics to medical research and historic preservation – there is a growing need for ef- ficient ways to browse and inspect large sets of high-resolution images. We present the development of a visualization software for solar physics data based on the JPEG 2000 image compression standard. Our implementation consists of the JHelioviewer client application that enables users to browse petabyte-scale image archives and the JHelioviewer server, which integrates a JPIP server, metadata catalog and an event server. JPEG 2000 offers many useful new features and has the potential to revolutionize the way high-resolution image data are disseminated and analyzed. This is especially relevant for solar physics, a research field in which upcoming space missions will provide more than a terabyte of image data per day. Providing effi- cient access to such large data volumes at both high spatial and high time resolution is of paramount importance to support scientific discovery. Introduction The Sun exhibits phenomena on all observable time scales and length scales, from seconds to tens of years, and from tens to hundreds of millions of kilo- 1 arXiv:0906.1582v1 [astro-ph.SR] 8 Jun 2009 meters. Over the last decade, the amount of data returned from space and ground-based solar telescopes has increased by several orders of magnitude. Space missions and ground-based observatories have been taking advantage of better optics, higher network capacities and greater storage capabilities to produce and deliver an ever-growing volume of solar data. This constantly increasing volume is both a blessing and a barrier: a blessing for making available data with significantly higher spatial and temporal resolutions, but a barrier for scientists to access, browse and analyze them. Today, the Solar and Heliospheric Observatory (SOHO1, launched in 1995) transmits about 200 MB of imagery per day. Its lineal descendant, the So- lar Dynamics Observatory (SDO2, to be launched at the end of 2009), will challenge scientists and engineers by sending back 1.4 TB of images per day. Among other data products, SDO will provide full-disk images of the Sun taken every 10 seconds in eight different ultraviolet spectral bands with a res- olution of 16 megapixel (MP) per image. This translates to 4096 x 4096 pixel resolution, or a single full image that no monitor or LCD display in the market today is large enough to display. With such staggering data volume, the data is bound to be accessible only from a few repositories and users will have to deal with data sets effectively immobile and practically difficult to download. From a scientist’s perspective this poses three problems: accessing, browsing and finding interesting data while avoiding the proverbial search for a needle in a haystack. An efficient solution for image encoding, storage and dissemination should therefore provide: • Remote access to images at different resolution levels, without increasing the already large data sets by storing multiple image resolutions; • Good compression performance to enable fast browsing while mitigating bandwidth bottlenecks; • Integrated viewing of image data and third-party metadata and event catalogs to easily locate data of interest; • Browsing capability for both still images and time series (movies). To achieve this goal with minimal additional storage requirements, it should be possible to generate movies on demand with a user-specifed resolu- tion, quality, field-of-view and time cadence; • Ability to handle, display and combine heterogenous data sets such as images from different sources with different image scales, resolutions and fields of view. The Jhelioviewer project aims at providing a solution to these challenges by putting a practical tool in the hands of those who are confronted with the enormous task of viewing and evaluating the increasingly greater volumes of 1http://soho.nascom.nasa.gov 2http://sdo.gsfc.nasa.gov 2 solar images. Its goal is to help scientists discover new phenomena and link related data sets from various instruments that are currently often analyzed in isolation. In addition, it will make a huge amount of information available

…(Full text truncated)…

This content is AI-processed based on ArXiv data.