The new Dimensionality Expansion tool creates additional bands in a multispectral image by computing second-order statistics and other nonlinear transforms on the original bands. Dimensionality expansion is a mathematical technique to increase multispectral data dimensionality in a nonlinear fashion so that standard hyperspectral linear methods can perform better at both pure-and mixed-pixel detection and classification. These hyperspectral methods include Orthogonal Subspace Projection (OSP), Constrained Energy Minimization (CEM), Matched Filter (MF), and Adaptive Coherence Estimator (ACE). The additional bands are created by computing the square roots, natural logs, auto-correlations (squares), and cross-correlations of the input bands.

The new Spectral Library Dimensionality Expansion tool runs dimensionality expansion on a spectral library. If you perform dimensionality expansion on image data, you must run the same expansion on any spectral library data you use.

The Download Web Data menu option lets you download remote sensing data from the web, extract the data, and open the data directly in ENVI.

The Reproject Raster tool has a simpler user interface.

The Layer Stacking tool has a simpler user interface with new options to define the output extent based on the geometric intersection or union of the input images.

In the Seamless Mosaic workflow, mosaics created from multispectral images now have a band-interleaved-by-pixel (BIP) format. This significantly reduces processing time. Mosaics of single-band images will continue to have a band-sequential (BSQ) interleave.

Reading MIE4NITF files is significantly faster, especially with large files.

New spectral filter functions are available for Sentinel-2A and -2B.

A new Duplicate Selected right-click option lets you copy and paste annotation items in a view.

You can delete multiple pixel-based and polygon-based ROI records by holding down the Shift key, then drawing a polygon around the records to remove.

ArcGIS Integration

ENVI Py 1.2 provides support for the ENVICOORDSYS data type. As a result, ENVIReprojectRaster can be deployed as a standalone tool in ArcMap and ArcGIS Pro. See the Running ENVI Analytics in ArcGIS Pro Tutorial to learn more.

ENVI Modeler Updates

A new Filter Iterator node executes an operation only when a specified condition is met. This node iterates through a collection of data while also applying the specified condition.

A Run Model in Debug Mode option was added to the ENVI Modeler menu bar. When the model runs, an Execution Report dialog reports the execution times for individual nodes and for the entire model. It also reports the values used for task parameters.

Multiple Aggregator nodes can be combined; for example, using one Aggregator node to close a loop introduced by an Iterator node, and using another Aggregator node to collect items into an array.

The documentation contains more examples of using Iterator and Aggregator nodes.

If a model produces text or numeric values, you can view those values in a dialog by connecting them to a View node.

NITF

The NITF/NSIF Module supports the following Tagged Record Extensions (TREs):

  • CSATTB and CSATTB_DESDATA: Common Sensor Attitude Data
  • CSCSDB and CSCSDB_DESDATA: Common Sensor Covariance Support Data
  • CSEPHB: Common Sensor Ephemeris Data
  • CSEXRB: Common Sensor Exploitation Reference Data
  • CSSFAB and CSSFAB_DESDATA: Common Sensor Field Alignment Data
  • CSSHPA: Common Sensor Shapefile
  • CSSHPB and CSSHPB_DES: Common Sensor Shapefile
  • CSWRPB: Common Sensor Warping Terms
  • MATESA
  • MTXFIL
  • RSMAPB: Replacement Sensor Model Adjustable Parameters
  • RSMDCB: Replacement Sensor Model Direct Error Covariance
  • RSMECB: Replacement Sensor Model Indirect Error Covariance
  • STREOA: Stereo Information

Programming

The ReprojectRaster task is no longer based on the ENVIReprojectRastervirtual raster.

New ENVITasks

Task

Description
BinaryAutomaticThresholdRaster Create an image where values above a specified threshold are set to 1 and all other values are set to 0. The task uses a predefined thresholding method to create the binary image.

CalculateGridDefinitionFromRasterIntersection

Create a grid definition from an array of rasters whose spatial extent encompasses the area where the rasters overlap.

CalculateGridDefinitionFromRasterUnion

Create a grid definition from an array of rasters whose spatial extent encompasses all of the rasters.

CalculateRasterThreshold

Calculate a threshold value for each band in a raster, as input to the BinaryGTThresholdRaster task or BinaryLTThresholdRaster task.

DimensionalityExpansionRaster

Create an image that contains an expanded number of bands created from various non-linear transforms. Having additional bands increases the dimensionality of a raster, which can help improve accuracy in some hyperspectral target detection and classification algorithms.

DimensionalityExpansionSpectralLibrary

Perform dimensionality expansion on a spectral library to match the data space of a dimensionality expansion raster.

ExtractROIsFromFile

Open one or more ROIs from a specified location.

PixelStatistics

Compute statistics for each pixel in a raster. The result is a new raster where each band represents a different statistic computed from the input raster.

New Routines

Routine

Description

ENVIDimensionalityExpansionRaster

Create an image that contains an expanded number of bands created from various non-linear transforms. Having additional bands increases the dimensionality of a raster, which can help improve accuracy in some hyperspectral target detection and classification algorithms.

Backward-Compatibility Notes

The EXPORT_AREA and EXPORT_CLASS_CLRS properties were removed from the ClassificationToShapefile task.