ENVI Feature Extraction Module (ENVI FX)

What is Feature Extraction?

Feature extraction – the process of finding and extracting specific objects of interest from high-resolution satellite imagery based on the object’s spatial, spectral, and texture characteristics – is an essential task for professionals across industries who use imagery for information. Many features can be identified and extracted from geospatial imagery such as vehicles, buildings, roads, coastlines, rivers, lakes, and fields. Once features are extracted, they can be used for many applications: in the military to find tanks or airstrips; in urban development to determine the extent of housing sprawl; or in local government to track highway assets.

Select the Image You Want to Use

Select an image from web or local search results, from ArcGIS, or from Windows® Explorer and simply drag and drop it into ENVI. ENVI supports today’s most popular satellite and airborne sensors, and it can extract features from panchromatic, multispectral, and hyperspectral data.

Segment Your Image

The ENVI Feature Extraction Module gives you the ability to extract a variety of features, depending on your specific needs and the imagery being analyzed. The feature extraction workflow guides you through the process of segmenting the image by dividing the image into discreet, real-world, objects through the identification of edges and the grouping of pixels into regions. A preview window provides immediate feedback on your segmentation without the need to process the entire dataset. This saves you time and provides the flexibility you need when experimenting with different results.

Enhance Your Data

If you have ancillary data about your image, such as a digital elevation model (DEM), you can import it into ENVI to provide additional criteria when segmenting and defining features. And, you can choose to confine image processing to specific areas of interest and reduce processing time.

Classify the Segments into Features

Following image segmentation, the workflow provides you with several options for defining the features that need to be extracted. You can choose example segments in the image that are representative of each feature class or create rule sets that must be met in order to assign segments to a feature class. You can also apply existing GIS or field data to supplement the classification and help ensure the most accurate parameters.

Refine the Results

Prior to final classification and extraction, the ENVI preview portal window allows you to view the preliminary results, so that you can make adjustments to the analysis before processing the entire image or example data. ENVI also provides tools to polish and refine your extraction results, obtain statistics, and assess accuracy so you can create a realistic final outcome. You can also choose to annotate results, count individual features, edit or smooth vectors, assess classification accuracy, and generate summary statistics about extracted features.

Share, Export, and Use Results

ENVI allows you to share results at all stages of the module, use results for further analysis or for reports, or save the results. Results can be printed, included as part of a PowerPoint presentation, or saved to your computer, a geodatabase, or directly into an ArcGIS file. Additional analysis can also be done using ENVI or ArcGIS.