Projects
Workbench for Imaging Spectroscopy Exploration and Research (WISER)
PI: Bethany Ehlmann (Division of Geological and Planetary Sciences)
SASE: Donnie Pinkston, Instructor
Imaging spectrometers in the visible/shortwave infrared range capture the majority of solar-reflected light, and add a spectral and thus compositional dimension to traditional spatial image data used in computer vision. These instruments are becoming more common in laboratory and field environments, where they enable rapid acquisition of quantitative maps of physical and chemical properties at high spatial resolution. They also have a long history of deployments on aircraft and spacecraft for remote mapping. They can be used to characterize terrestrial and coastal aquatic ecosystems, geology, and atmospheric properties, and are also critical tools for exploring other planetary bodies.
Unfortunately, to this point, the most powerful software for viewing and analyzing imaging spectroscopy data is only available under a commercial license with significant fees attached, even when used within academic or research settings. This reduces the opportunity for researchers and students to explore and analyze this valuable data resource.
This Schmidt Academy project has focused on developing a new graphical application called WISER – Workbench for Imaging Spectroscopy Exploration and Research – that is completely free for use in academic and research settings. (Commercial use of WISER can be arranged through the Caltech Office of Technology Transfer.) After approximately two years of development, WISER provides a solid foundation of features for viewing and analyzing imaging spectroscopy data, with features like the following:
- Highly configurable three-level viewing hierarchy for exploring data, including a context window, a main image window, and a zoom window
- Main image window can be subdivided into multiple panes, for viewing multiple images concurrently
- Linked-image viewing, e.g. for viewing a mineral sample alongside a computed mineral map
- Support for geographic coordinate reference systems, showing geographic coordinates of UI pixel selections, and allowing users to navigate to specific geographic coordinates in an image
- User-specified regions of interest comprised of rectangles, polygons, and multi-pixel selections
- Spectrum viewing window to show spectrum-at-point waveforms, mean/median spectra of regions around points, average spectra of regions of interest, spectral-library data, etc.
- Rich band-math capabilities for evaluating a wide range of math expressions involving image cubes, image bands, spectra, etc.; results may be visualized directly within WISER
- Plugin API for extending WISER with custom graphical tools
- Plugin API for extending WISER band-math with custom analysis functions
- Ability to import and export individual spectra, or spectra from regions of interest
- Ability to import and export regions of interest
Future work on WISER will continue to grow out these features, and to develop a rich library of plugins useful for researchers and students.
More information available at http://wiser.caltech.edu
[Fig. 1] Users can perform ad-hoc analysis of spectral imaging data with band-math expressions on images, bands and spectra."
[Fig. 2] Images with geographic coordinate reference systems allow WISER to show and/or navigate to the geographic coordinates of image features.
[Fig. 3] Related images may be linked in the UI, so they zoom/scroll together, and share a unified set of selections.
[Fig. 4] Users can create Regions Of Interest (ROIs) from rectangles, polygons, and multi-pixel selections, then perform operations on them like viewing the mean/median spectrum from the ROI.