ENVI 4.5 Product Updates

1
ENVI 4.5 Product Updates
2
ENVI 4.5 Value Proposition

• ArcGIS Interoperability Geodatabase and ArcMap
  access
• Fx enhancements Improved end-to-end workflow
• New analytical spectral tools SID and ICA are
  powerful tools for image scientist
• Continued data format support Worldview, NITF
  and TFRD updates

3
ENVI and ArcGIS Interoperability

• ENVI users will be able to

• Access and read from ArcGIS Geodatabases

• Write out to geodatabases

• Launch ArcMap and create map compositions
  directly from ENVI

• Supports enterprise, personal, and file
  Geodatabase types

4
New Sensor and Data Support

• WorldView-1 data in GeoTIFF, NITF, and Stereo
  Pair.
• NITF for 32- and 64-bit Linux systems
• Full TFRD support in ENVI Zoom
• Remote Data Access in ENVI (OGC/JPIP), not just
  in FX

5
New Feature Extraction Functionality

• Programmatic access to Fx provides batch mode
  processing for production facilities and users
  with ENVI/IDL expertise
• envi_doit, envi_fx_doit
• Vector and Attribute Export allows customers to
  use Fx segmentation results in other analyses and
  workflows
• export segments as features prior to
  classification

6
New Feature Extraction Functionality

• Summary Panel provides quick, practical, portable
  results

7
New Feature Extraction Functionality

• Vector Smoothing for cleaner and smoother
  results, with fewer vertices and generalized
  features
• Douglas-Peucker based algorithm
• works on line and polygon output
• user controls smoothing threshold in pixels
• Create new vectors directly in ENVI Fx

8
New Feature Extraction Functionality

• Save/Restore training data for supervised
  classification
• Reuse training sets on libraries of imagery
• Restore multiple training sets to build
  multi-source feature library
• Use saved training data as input to batch Fx

9
New Spectral Algorithm Independent Component
Analysis (ICA)

• ICA transforms a set of mixed random signals into
  components that are as mutually independent as
  possible
• Independent components uses higher order
  statistics than Principal Components. Better
  distinguishes weak signals from background noise.

10
Independent Component Analysis

• ICA can be used to find and separate hidden
  (noise) component

11
Independent Component Analysis

• Other applications of ICA
• anomaly detection
• dimension reduction
• noise reduction
• classification
• endmember extraction
• Programmatic interface
• envi_doit, envi_ica_doit
• envi_doit, envi_ica_inv_doit

12
New Spectral Algorithm Spectral Information
Divergence (SID)

• Classification technique (similar to SAM) that
  uses a divergence measure to match pixels to
  reference spectra
• SID can characterize spectral similarity and
  variability more effectively than SAM
• As with all spectral tools, programmatic access
  is provided

13
ENVI 4.5 with IDL 7.0

• A single, cross-platform IDL development
  environment that looks and behaves like a native
  application
• The foundation for new functionality that will
  continue to improve the usability of ENVI IDL