X

Help Articles are product support tips and information straight from the NV5 Geospatial Technical Support team developed to help you use our products to their fullest potential.



5820 Rate this article:
No rating

Info About 'Use Subspace Background' Option in Target Detection Wizard

In step 6 of ENVI's Target Detection Wizard (Select Target Detection Methods) if you click on the Show Advanced Options button, you will see an option for 'Use Subspace Background.' If you select this option, a Background Threshold parameter will appear. The documentation describes this option in this way:

"If you selected the ACE, CEM, MF, TCIMF, MTMF, and/or MTTCIMF method(s), you can model the scene background by removing anomalous pixels before calculating background statistics. To model the subspace background, enable the Use Subspace Background check box. By better characterizing the subspace background, you achieve greater target-to-background separation and can potentially improve ACE, CEM, MF, TCIMF, MTMF, and/or MTTCIMF detection results, particularly in scenes that contain a lot of clutter or man-made objects.

If Use Subspace Background is enabled, specify the Background Threshold to use for calculating the subspace background statistics. This value indicates the fraction of the background in the anomalous image to use. The allowable range is 0.500 to 1.000 (the entire image)."

It may be useful to know that this process involves calculating an anomaly image using the RxD method (Chang, C. and Chiang, S., 2002). The areas detected as anomalous are then excluded from the calculation of the background statistics, to remove the influence of anomalous features from the characterization of the general background.

It may also be helpful to know that setting the threshold to 1.0 means the entire image will be used to calculate the background stats. Setting the threshold to 0.8 means that the 80% of pixels that have lowest anomaly values will be used to calculate the background stats.

--------------------------------

Chang, C. and Chiang, S. 2002. Anomaly detection and classification for hyperspectral imagery. IEEE Transactions on Geoscience and Remote Sensing, 40(6):1314-1325.

Please login or register to post comments.
Featured

End-of-Life Policy Enforcement for ENVI 5.3 / IDL 8.5 and Earlier Versions

5/6/2024

April 1, 2024 Dear ENVI/IDL Customer,  We are reaching out to notify you of our supported... more »

How to Upgrade licenses to ENVI 6.x / IDL 9.x

12/5/2023

What is the new Upgrade function? Starting with ENVI 6.0 and IDL 9.0, we have implemented an... more »

What to do if the 'License Administrator - License Server' for the Next-Generation License Server does not start?

6/13/2023

Background: With the release of ENVI 5.7 & IDL 8.9 and the corresponding Next-Generation licensing... more »

Next-Generation Licensing FAQ

4/28/2023

  NV5 Geospatial has adopted a new licensing technology for all future releases of our ENVI, IDL... more »

The IDL Virtual Machine

6/6/2013

What is the IDL Virtual Machine? An IDL Virtual Machine is a runtime version of IDL that can... more »