SOM_VIS is a computer program for analysis and display of multidimensional sets of Earth-image data typified by the data acquired by the Multi-angle Imaging Spectro-Radiometer [MISR (a spaceborne instrument)]. In SOM_VIS, an enhanced self-organizing-map (SOM) algorithm is first used to project a multidimensional set of data into a nonuniform three-dimensional lattice structure. The lattice structure is mapped to a color space to obtain a color map for an image. The Voronoi cell-refinement algorithm is used to map the SOM lattice structure to various levels of color resolution. The final result is a false-color image in which similar colors represent similar characteristics across all its data dimensions. SOM_VIS provides a control panel for selection of a subset of suitably preprocessed MISR radiance data, and a control panel for choosing parameters to run SOM training. SOM_VIS also includes a component for displaying the false-color SOM image, a color map for the trained SOM lattice, a plot showing an original input vector in 36 dimensions of a selected pixel from the SOM image, the SOM vector that represents the input vector, and the Euclidean distance between the two vectors.
This program was written by P. Peggy Li, Joseph C. Jacob, Gary L. Block, and Amy J. Braverman of Caltech for NASA's Jet Propulsion Laboratory. For further information, access the Technical Support Package (TSP) free on-line at www.techbriefs.com/tsp under the Software category.
This software is available for commercial licensing. Please contact Karina Edmonds of the California Institute of Technology at (818) 393-2827. Refer to NPO-40666.
This Brief includes a Technical Support Package (TSP).
Self-Organizing-Map Program for Analyzing Multivariate Data
(reference NPO-40666) is currently available for download from the TSP library.
Don't have an account?
Overview
The document outlines the development of a visualization and analysis system called SOM_VIS, which utilizes an enhanced Self-Organizing Map (SOM) algorithm combined with a Voronoi Refinement algorithm. This system is designed to effectively represent multi-dimensional image datasets, specifically targeting data from the Multi-angle Imaging SpectroRadiometer (MISR). The primary goal of SOM_VIS is to visualize multispectral and multivariate Earth Observing System (EOS) datasets while preserving the spatial relationships inherent in the original data.
The process begins with the application of the SOM algorithm to project high-dimensional datasets into a non-uniform 3D lattice structure. This lattice is then mapped to a color space, creating a colormap for the image. The Voronoi cell refinement algorithm enhances the mapping of the SOM lattice to various levels of color resolution, resulting in a false color image where similar colors indicate similar characteristics across multiple data dimensions.
SOM_VIS includes several user-friendly components: a data control panel for selecting subsets of MISR’s Level 1B2 Radiance data products, a training control panel for adjusting parameters for SOM training (such as lattice size and convergence speed), and a display for the resulting false color SOM image. Additionally, it provides a colormap for the trained SOM lattice and a plot illustrating the original input vector in 36 dimensions.
The document also discusses the patent status of the technology, indicating that it is related to assigned work and has not yet been published. Plans for future publication and presentations at conferences, such as the AGU Fall Meeting in December 2003, are mentioned, highlighting the ongoing interest in the technology.
The SOM_VIS system has been developed as a prototype and has been demonstrated to the MISR team and NASA management. While it is fully functional, further development is needed to generalize the system for other EOS datasets and to incorporate additional features. The technology is expected to significantly enhance the understanding of relationships between Level 1 radiance data and Level 2 derived geophysical products, ultimately benefiting JPL’s Earth Science missions and programs.
In summary, SOM_VIS represents a significant advancement in the visualization of complex multi-dimensional datasets, with potential applications in various scientific fields, particularly within the NASA community.
