Visualization and logical binding of hyperspectral data using QuickViz and Saada

Petremand, Matthieu

Next generation integral-field spectrographs such as MUSE will acquire large hyperspectral cubes and thus generate a huge amount of data (up to 1.2 GB for a single MUSE observation) whose visualization remains a problematic and crucial task before any processing. In addition, large observation parameters such as variance values, PSF widths or data quality flags may also be associated to each cube and involved in the scientific interpretation of the data.

In the framework of the DAHLIA project, we propose a new software, QuickViz, providing a set of basic and advanced features enabling easy exploration of such hyperspectral datasets: synchronized visualization between both spatial and spectral domains, customizable visualization modes and algorithms, data extraction and comparison, variance display, multi-view mode... QuickViz is designed in Java as an Aladin plug-in and thus extends its hyperspectral functionalities while managing all the interactions between cubes and extracted spectra. An optimized memory management together with an easy-to-use interface allow to quickly browse and compare several large hyperspectral images whereas the finer astronomical analysis can be delegated to VO (Virtual Observatory) tools thanks to the IVOA SAMP protocol.

Large data files, the necessity to bind extra information to support data interpretation, this partly covers the same requirements stated for the development of the SAADA application. An illustration of a SAADA repository housing hyperspectral cubes linked with local dispersion matrix and new features of SAADA will be shown as well.

Return to demo list