|
|
|
|
| The ENVI Fx Step-by-Step Workflow |
| |
|
Find features in your geospatial imagery quickly and easily.
|
|
You'll find even more automated workflow tools in the ENVI Feature Extraction Module (ENVI Fx) to quickly, easily and accurately extract features of interest from your geospatial imagery. ENVI's automated workflows provide step-by-step processes and instructions to help you get the information you need from your imagery.
|
|
|
|
Prepare Your Data
As you initiate the feature extraction workflow, ENVI Fx prompts you with an input dialog box. This step allows you to select a base image and ancillary data so you can get comprehensive information about your area of interest.
-
Define a panchromatic or multispectral base image for the primary feature extraction workflow
-
Incorporate ancillary data sets, such as elevation models, to provide additional criteria when segmenting and defining features
|
|
-
Ingest files directly from your file system, from a geodatabase, from any OGC WCS server, or from any JPIP server
-
Include a mask to confine processing to restricted areas of interest and reduce the overall processing time
|
|
|
Segment Your Image
Following data preparation, ENVI Fx guides you through the process of segmenting your image. In this step, ENVI Fx divides the image into discreet, real-world objects by finding edges and grouping pixels into regions.
-
Scale-based approach allows you to extract features of various contrasts and sizes
-
Compute spatial, spectral and textural attributes to describe the physical properties of the segments and prepare for the segment definition step
|
|
|
|
|
Define Your Segments
With segmentation complete, ENVI Fx provides you with several options for defining features to be extracted. At this point in the workflow, you have the option to export your segmentation as vectors, or perform supervised or rule-based classification. With either method, you can save your feature definitions for later, share results with colleagues, or create feature libraries.
|
Export vectors
The ENVI Fx workflow allows you to export your segmentation results as vectors, along with all of their computed attribute information, to use in your own analyses, such as an input into an ArcGIS Geoprocessing Model
Import ground truth data
Use exiting GIS or field data to supplement your classifications
Classify by creating rules
Define a set of conditions that must be met in order to assign segments into a feature class
|
|
Classify by using examples
Pick or identify segments in the image that are representative of each feature class you define. ENVI Fx will use this information to identify similar segments and place them in the matching feature class
Preview results
Prior to final classification, you can preview the results and alter your rules before processing the entire image or example data
Modify
The classification algorithms available are flexible to allow for custom modifications to accommodate specific classification needs
|
|
|
Refine Your Results
ENVI Fx provides tools to polish and refine your extraction results, obtain statistics, and assess accuracy so you can create a realistic final outcome.
|
Annotate extracted results to create easily understood maps and reports; a full suite of annotation tools is included and customizable
Count individual features across an image scene with an easy-to-use, interactive tool
Smooth vectors and adjust results to the proper level of detail in order to make objects more realistic
|
|
Edit vectors by manipulating individual vertex points or modifying general properties of the vector layer as a whole
Assess classification accuracy visually using display portals to compare features against the source imagery
Generate summary statistics of extracted features, including the number of features in each class and total feature area
|
|
|
Save, Repeat, or Share the Outcome
ENVI Fx allows you to share results at all stages of the workflow. Easily use the results for further analysis, for reports, or to share with colleagues.
-
Easily save vector results to a geodatabase for use in subsequent analyses in ENVI or for use with other products
-
Combine multiple types of data, such as LiDAR or elevation data, with your input data source for increased accuracy
|
|
-
Save training data, classifications, and rules sets to create a library for use in subsequent workflows or to share with colleagues
-
Employ a programmatic method for batch processing so you can automatically extract features from multiple datasets simply by setting parameters and specifying an output directory
|
|
|
|
|
|
|
|