ENVI Glossary - ENVI Tutorial
ENVI Glossary 2 (O - Z)

A region of interest with spatial, spectral, and/or texture characteristics (brightness, color, etc.) that define the region.

object-based processing

techniques that classify a set of input objects rather than classifying pixels individually.


In the context of remote sensing:

  • A correction value added to or subtracted from every pixel in an image, typically by using image arithmetic (Band Math in ENVI).
  • A variable added to the gain in a regression equation for sensor calibration.
  • The number of bytes of embedded header information present in the file (ENVI skips these bytes when reading the file).

Operational Linescan System; a NOAA DMSP data format.


To remove the effects of radial relief displacement and imaging geometry from remote sensing imagery.


A sensor that detects electromagnetic energy in one very broad band, which includes most of the visible light spectrum. In aerial photography, panchromatic refers to a type of film that is sensitive to all wavelengths of visible light.


Planetary Data System data format.

pedestal height

The height of a polarization signature (above 0), calculated by averaging the following four polarization combinations from SAR data:

Table 9-1: Polarization Combinations










phase image

An image derived from polarimetric SAR data whose values represent the phase difference between the horizontal and vertical polarizations. The phase difference is measured in either radians or degrees and ranges from -π to π, or -180 degrees to 180 degrees.


Windows QuickDraw Picture format.

pixel-based processing

The traditional approach to exploitation and classification where each pixel in an image is treated as an independent data point.

Pixel Purity Index

A tool used to find the most spectrally pure (extreme) pixels in multispectral and hyperspectral images. These typically correspond to mixing endmembers. The PPI is computed by repeatedly projecting n-D scatter plots on a random unit vector. ENVI records the extreme pixels in each projection (those pixels that fall onto the ends of the unit vector) and it notes the total number of times each pixel is marked as extreme. A Pixel Purity Image is created where each pixel value corresponds to the number of times that pixel was recorded as extreme.

plot function

In ENVI programming, a user function that you can add to and call from the Plot_Function menu of any ENVI plot window.


Portable Network Graphics image format.

polarization signature

A plot of radar backscattered power as a function of ellipticity and orientation angles of the incident radar wave.


A continuous line composed of one or more segments; a vector and annotation object.

polynomial function

A mathematical function with the following form:

f(x) = anxn + an-1xn-1 + ... + a1x + a0


n is a nonnegative integer

an, an-1, etc. are coefficients.

The degree of the polynomial function is the highest value for n, where an is not equal to 0.

principal components analysis

A mathematical technique that transforms a multivariate data set into a new coordinate system such that the axes, or principal components, of the new coordinate system are uncorrelated. In remote sensing, an image is created for each principal component. Because the principal component rotation maximizes the variance in the first few principal components bands, these bands usually contain most of the coherent image information and can be used to isolate features in the data.


In ENVI programming, a sequence of one or more IDL statements that you can assign a name (thus creating an IDL program), compile, and call from the IDL or ENVI command line, using the following form:

PROCEDURE_NAME, Argument [, Optional_Argument]   
projected coordinate system

A coordinate system that uses Cartesian coordinates (x,y) to describe a geographic location.

pushbroom sensor

Also called an along-track scanner; a sensor with a line array of small, sensitive detectors stacked side-by-side, where each detector corresponds to a pixel in the resulting image. As the satellite advances along the ground track, the array of detectors receives radiation simultaneously. Examples of pushbroom sensors include ASTER, IKONOS, OrbView-3, QuickBird, SPOT, and CARTOSAT-1.

pyramid layers

Copies of a data set at various reduced resolutions. They are used to speed image display by reducing the resampling required when displaying large portions of an image at low resolution.


Quick Atmospheric Correction; an automated atmospheric correction method in ENVI for retrieving spectral reflectance from multispectral and hyperspectral images.


A Digital Globe high-resolution satellite that provides 61 cm panchromatic and 2.4 m multispectral imagery.


An ENVI feature that allows you to quickly create a map composition from an image. You can add grid lines, scale bars, titles, north arrows, declination diagrams, and logos. You can save your settings as a QuickMap template that you can use with other images.


A type of receiving station that creates AVHRR 16-bit High Resolution Picture Transmission (HRPT) files with two header frames. The Quorum format does not have georeferencing information.


Radar Satellite (Canadian Space Agency).


A measure of the amount of electromagnetic radiation leaving a point on the surface. More precisely, it is the rate at which light energy is emitted in a particular direction per unit of projected surface area. The standard unit is W/m2. Most remote sensing devices directly measure radiance.


A grid-based data structure for storing images where each cell, or pixel, contains a single data value.


The ratio of radiant energy reflected by a body to the energy incident on it, usually denoted as a percentage.


A broad term that refers to a group of pixels with the same spatial or spectral characteristics.


To geometrically align two or more images of the same scene so the images can be superimposed. The images can come from different viewpoints, different times, and different sensors. Following are the two most common methods of registration:

  • Image-to-image registration: Correct a reference (warped) image to match base image geometry, using tie points between the two images.
  • Image-to-map registration: Assign a reference image to geographic coordinates, using GCPs with known ground locations.

To apply a geometric transformation to an original data set; more specifically, the interpolation method used to derive output pixel values based on input pixel values, taking into account the computed distortion. Following are the most common resampling methods:

  • Nearest neighbor: Each pixel in the output image receives its value from the nearest pixel in the input (reference) image.
  • Bilinear: Each estimated pixel value in the output image is based on a weighted average of the four nearest neighboring pixels in the input image.
  • Cubic convolution: Each estimated pixel value in the output image is based on a weighted average of 16 nearest neighboring pixels in the input image. Cubic convolution is the slowest method, but it yields the smoothest results.

A color space defined by red, green, and blue values.

RGB color composite

An image that uses the red, green, and blue guns of the display device to form a color additive representation of color.

ROC curve

Receiver operating characteristic; a curve used to visualize the performance of a classification method, in order to select the proper decision threshold. ROC curves compare a series of rule image classification results for different threshold values with ground truth information.


Region of interest; a point, polyline, or polygon object drawn on an image, used to define a specific area of interest for extracting classification statistics, masking, and other operations in ENVI. From a processing standpoint, ROIs are pixel addresses with associated data.

RMS error

Root mean square error; a statistical measure that represents the difference between measured and predicted data points. In ENVI, RMS error is often used to evaluate a set of GCPs for georeferencing:

ENVI Gloassay - facegis.com


x and y are the original row and column coordinates

x' and y' are the estimated coordinates


A general IDL programming term that refers to both functions and procedures.


Rational polynomial coefficients; used to build interior and exterior orientation in photogrammetry.


Replacement sensor model; an alternate representation of sensor geometric information that corrects the deficiencies of RPC-based sensor models. RSM contains a variety of enhancements over the RPC model, including:

  • Increased accuracy over images with large number of rows or columns (such as image strips) by breaking the image into tiles with separate models.
  • The ability to store varying degrees of complexity in the polynomial representation used.

Rotation, scaling, and translation; a warping method used in image registration that uses an affine transformation with at least three ground control points:

x = a1 + a2X + a3Y

y = b1 + b2X + b3Y

rule image

An image calculated for each ROI in a supervised classification. They are called "rule images" because a rule is applied to the pixel values in the images to determine the class to which each pixel should be assigned. The pixel values in the rule images, and the rule used to assign classes, depend on the specific classifier used. For example, with Maximum Likelihood classification, the pixel values in a rule image for one class are equal to the likelihood that each pixel belongs to that class. Whichever rule image has the highest likelihood value for a pixel is the class to which that pixel is assigned.


Reed-Xiaoli anomaly detection algorithm.


The x component of a raster image coordinate pair (x,y); same as column.


Synthetic aperture radar.

scale factor

A division factor used to convert integer-scaled reflectance or radiance data into floating-point values. For example, for reflectance data scaled into the range of 0 to 10,000, set the scale factor to 10,000. For uncalibrated integer data, set the scale factor to the maximum value the instrument can measure ((2n) - 1, where n is the bit depth of the instrument).

scatter plot

A plot of measurements from two or more bands of data.

Scroll window

The window in a display group that displays the full image at subsampled resolution. This window appears only when an image is larger than what ENVI can display in the Image window at full resolution.


Sea-viewing Wide Field-of-view Sensor; a NASA satellite that collects global ocean color data.


The process of partitioning an image into connected regions by grouping neighboring pixels with similar feature values (brightness, texture, color, etc.). These segments ideally correspond to real-world objects.


A statistic that uses the squared difference between neighboring pixel values to provide a measure of dissimilarity within a dataset. It has the same units as the input dataset, and its values are greater than or equal to 0.


A plot of semivariance values at multiple lag distances; a measure of how autocorrelation decreases as distance increases.

shaded relief

An image created during the topographic modeling process that renders terrain in 3D by use of graded shadows that would be cast by the sun from a northwest direction.


A vector file format; a set of files that contain points, arcs, or polygons that hold tabular data and a spatial location. One shapefile consists of three individual files, ending with .shp, .shx, and .dbf file extensions.

shift difference

The process of differencing adjacent pixels to the right and above each pixel and averaging the results to obtain the noise value to assign to the pixel being processed. The best noise estimate is gathered using the shift-difference statistics from a homogenous area rather than the whole image.

Sigma nought

A measure of the mean backscatter of a radar signal from an area of 1 m2 on the earth's surface, typically denoted in decibels (dB). Sigma nought describes the backscattering strength of a distributed target, rather than a discrete target.


A synthetic aperture radar (SAR) instrument built by NASA/JPL and Ball Communication Systems Division for NASA. SIR-C flew aboard the Space Shuttle and provided L-band and C-band measurements.


The percentage or degree change in elevation over distance; a parameter used in topographic modeling.


Sequential Maximum Angle Convex Cone; a sub-pixel spectral tool in ENVI that finds spectral endmembers and their abundance throughout an image.


To average pixel values within adjacent areas to produce more gradual transitions.

solar azimuth

The angle a horizontal projection of a direct ray from the sun makes with the True North-south axis, typically denoted as clockwise from True North through 360 degrees.

solar elevation

The angle of the sun above the horizon, extending from 0 degrees (horizon) to 90 degrees (directly overhead).

solar spectrum

The part of the electromagnetic spectrum occupied by the wavelengths of solar radiation. About 99 percent of solar radiation is constrained to 300 nm (ultraviolet) to 3,000 nm (near-infrared).

spatial filter

A filter that removes certain spatial frequencies from an image and enhances features in the remaining image. Following are the most popular types of spatial filters:

  • High pass: Enhances high spatial frequencies. Following is a sample 3 x 3 kernel used for high-pass filters:
  • -1 -1 -1

    -1 9 -1

    -1 -1 -1

  • Low pass: Enhances low frequencies in an image, thus smoothing the image. Following is a sample 3 x 3 kernel used for low-pass filters:
  • 1 1 1

    1 1 1

    1 1 1

  • Edge enhancement: Enhances edges of features in an image, making them easier to analyze. Following is a sample 3 x 3 kernel used for edge enhancement filters:
  • 1 1 1

    1 -2 1

    -1 -1 -1

spatial profile

A plot of pixel values along a line placed in the image.

spatial resolution

A measure of the smallest angular or linear separation between two objects that a sensor can resolve.

Spectral Analyst function

In ENVI programming, a user function that implements a custom spectral mapping method to match an unknown spectrum to the materials in a spectral library. You can add this user function to ENVI's Spectral Analyst and call it from the ENVI menu system.

spectral library

A collection of spectra measured in the field or laboratory for materials (minerals, vegetation types, etc.) that are often used as a baseline, or "true" spectra, for identification of materials from spectral remote sensing imagery.

Spectral mapping method

Also called a spectral similarity technique; a method in hyperspectral analysis for matching image spectra to known (reference) spectra, usually from a spectral library. Following are descriptions of the spectral mapping methods used in ENVI.

  • Binary Encoding: A method that encodes data and endmember spectra into zeros and ones, based on whether a band falls below or above the spectrum mean, respectively.
  • Linear Spectral Unmixing: A sub-pixel method that determines the relative abundance of materials depicted in multispectral or hyperspectral imagery based on the materials' spectral characteristics.
  • LS-Fit: A linear band prediction method that uses least-squares fitting. You can use it to find regions of anomalous spectral response in a dataset. It calculates the covariance of the input data and uses it to predict the selected band as a linear combination of the predictor bands plus an offset.
  • Matched Filtering: A method that finds the abundance of user-defined endmembers using a partial unmixing technique. Matched filtering maximizes the response of a known endmember and suppresses the response of the unknown background, thus matching the known signature. It provides a rapid means of detecting specific materials based on matches to library or image endmember spectra and does not require knowledge of all the endmembers within an image scene.
  • Mixture Tuned Matched Filtering (MTMF): A method that performs Matched Filtering and adds an infeasibility image to the results. The infeasibility image is used to reduce the number of false positives that are sometimes found when using Matched Filtering.
  • Spectral Angle Mapper (SAM): A physically-based spectral classification method that uses an n-D angle to match pixels to reference spectra. The algorithm determines the spectral similarity between two spectra by calculating the angle between the spectra and treating them as vectors in a space with dimensionality equal to the number of bands.
  • Spectral Feature Fitting (SFF): A method that compares the fit of image spectra to reference spectra using a least-squares technique. SFF is an absorption-feature-based methodology. The reference spectra are scaled to match the image spectra after the continuum is removed from both data sets.
Spectral Math

An ENVI tool that allows you to apply mathematical expressions or IDL procedures to spectra and to selected multiband images, as long as the number of bands and spectral channels match.

spectral profile

see Z Profile.

Spectral Hourglass Wizard

A tool in ENVI that takes you through a step-by-step process for locating spectral endmembers within a hyperspectral data set and mapping their locations and sub-pixel abundances. When written in a certain format, the processing flow resembles an hourglass shape.

spectral resolution

The wavelength range that a particular band measures. For example, Landsat-7 ETM+ Band 1 detects wavelengths from 0.45 μm to 0.52 μm. The Landsat-7 ETM+ panchromatic band detects wavelengths from 0.50 μm to 0.90 μm. So, Band 1 has a finer spectral resolution than the panchromatic band. Spectral resolution does not refer to the number of bands available from a particular sensor.


Satellite Pour l'Observation de la Terra; a series of earth observation satellites launched by Spot Image of France; the spatial resolution of SPOT data varies from 2.5 to 20 m.


Spectral response function; engineering data that quantify the spectral response and sensitivity of detectors on an airborne or satellite sensor. The term SRF also refers to a data format that contains SRF data for particular sensors.


Shuttle Radar Topography Mission (NASA/JPL).


Numbers or values that help describe the characteristics of a selected group of data.


See contrast stretch.


To create an output image of reduced resolution based on every nth pixel from an input image.

super GLT

A geographic lookup table (GLT) file that contains information about how many and which input pixels contribute to the output pixel; a super GLT is not an image file.


With respect to polarimetric radar data, the process of building an image representing backscatter at specified ellipticity and orientation angles from a scattering (or related) matrix.

system calibration

The process of converting digital numbers (DNs, which represent the sensor response) in a remote sensing image to radiance or reflectance above the atmosphere, using pre-launch gain and offset values.

temporal resolution

How often a sensor obtains imagery of a particular area.


The frequency of change and spatial arrangement of pixel values in an image, as a function of spatial scale. A flat image in which all digital numbers (DNs) are equal is said to have a uniform texture.

texture filter

A filter used to delineate surface features (for example, biophysical properties of a forest canopy) that cause local variations in image brightness. A texture filter is helpful for identifying objects that are more characterized by their texture than by intensity.

thermal band

A band that detects radiation from the far infrared part of the electromagnetic spectrum, between approximately 7.0 to 15 μm. A thermal band detects emitted, rather than reflected, radiation from the earth.


A value above which a process is performed and below which it will not be performed. For example, you can specify a change threshold of 50 percent in a K-Means supervised classification, which means the clustering process ends when the number of pixels in each class changes by 50 percent or less.


In the Spatial Feature Extraction module, thresholding refers to the rules and limits for distinguishing objects from the background, based on local pixel information. Thresholding is an optional step used to extract point features (for example, airplanes) when over-segmentation is a problem. This step groups adjacent segments into new segments based on the DN values of each segment.

tie points

The location of a single feature across two overlapping images, used in image-to-image registration.


Tagged Image File Format.


For ENVI to process images that are much larger than the total amount of RAM available on the system, large images are broken into pieces that are small enough for the system to handle. Each piece is called a tile. When the image is processed, only one tile is read into memory at a time.


Thermal Infrared Multispectral Scanner; a NASA/JPL and Daedalus Enterprises sensor.


Thematic Mapper; a Landsat sensor.


Topographic Synthetic Aperture Radar; a NASA/JPL airborne radar interferometer.


An image processing operation that changes data to another data space, usually by applying a linear function. The goal of most transforms is to improve the presentation of information. Transformed images are often more easily interpreted than the original data.

true-color composite

An image where the red band (0.6 μm to 0.7 μm) is displayed in red, the green band (0.5 μm to 0.6 μm) is displayed in green, and the blue band (0.4 μm to 0.5 μm) is displayed in blue.

user-defined RPC reader

In ENVI programming, a user function that reads a custom rational polynomial coefficients (RPC) file format. You can add the user function to, and call it from, the Generic RPC menu option in ENVI.

user function

A custom program that you can write in IDL, C, Fortran, or another high-level programming language that performs an interactive ENVI processing task. If you write a user function in IDL, you can incorporate ENVI library routines into the code and call the user function from the ENVI menu system to run it.


United States Geological Survey.


Uniform Target Detector; an anomaly detection algorithm in which the anomaly is defined using (1 - μ) as the matched signature, rather than (r - μ). UTD and the RXD algorithm work exactly the same, but instead of using a sample vector from the data (as with RXD), UTD uses the unit vector. UTD extracts background signatures as anomalies and provides a good estimate of the image background.


Universal Transverse Mercator map projection; a system of plane coordinates based upon 60 north-south zones, each 6 degrees of longitude wide, that circle the earth. UTM coordinates consist of the Zone number, an easting (distance in meters or kilometers east of the western edge of the Zone), and a northing (distance in meters or kilometers from the equator).


A data structure for storing spatial data that consists of points, lines, and polygons. Lines (also called arcs) are defined by beginning and end points, which meet at nodes. The locations of these nodes and the topological structure are usually stored explicitly.


Also called object generation; In the Spatial Feature Extraction module, vectorization refers to the process of creating vector objects from segmentation results. ENVI Zoom performs this step in the background without user input.

vegetation index

A measure of some vegetation property calculated from reflected solar radiation measurements made across the optical spectrum (400 to 3,000 nm). A vegetation index is constructed from reflectance measurements in two or more wavelengths to analyze specific characteristics of vegetation, such as total leaf area and water content. See ENVI Help for the names and definitions of each vegetation index calculated in ENVI.

vertical exaggeration

The process of multiplying all of the elevation values in a DEM by a scale factor to exaggerate the landscape's relief when viewed in a 3D perspective.

virtual border

An annotation object that consists of a temporary border around an image displayed in ENVI. You can enter the border width (in pixels) and place other annotation objects within the virtual border.

virtual Mosaic

A saved mosaic template used as an alternative to saving a mosaic to disk. When you restore a Virtual Mosaic template file, ENVI opens the individual image files that make up the mosaic and puts them together on the fly. You can display a Virtual Mosaic file in ENVI and annotate it, stretch it, etc., like any other image. Using a Virtual Mosaic prevents multiple files from containing the same images and therefore saves disk space.


To stretch an image to fit its ground control points (GCPs), so that distance and area are uniform in relationship to real-world measurements. ENVI performs warping with rotation, scaling, and translation (RST); polynomial, or Delaunay triangulation.


Velocity divided by frequency of an electromagnetic wave. In general, the mean distance between maxima or minima of a roughly periodic wave pattern.


A simple graphical object such as a push button or slider, created in IDL, that allows user interaction with a pointing device (usually a mouse) and a keyboard. You can construct and manipulate graphical user interfaces in IDL using widgets.


A Digital Globe high-resolution satellite that provides .5 m panchromatic imagery.

X Profile and Y Profile

A cross-section of data along the x-axis, and y-axis, of an image, respectively. X and Y Profiles are also called horizontal and vertical profiles, respectively.


X Windows Dump


Variables in an image header file that define the image coordinates for the first pixel in the image. For most images, ENVI sets the default XSTART and YSTART values to 1, defining the first pixel in an image with a coordinate of (1,1). Thus, if the image were an IDL 2D array variable, the data contained in subscript position [0, 0] correspond to image coordinates (1,1). If XSTART or YSTART are set to any other values (including negative numbers or 0), the image coordinates begin incrementing from these values.

Z Profile

A spectrum plot of the pixel under the cursor, through all bands of the image.

Zoom box

The red box inside an Image window that defines the extent of the Zoom window.

Zoom window

The window in a display group that displays the subsection of the image defined by the Image window Zoom box. The resolution is at a user-defined zoom factor based on pixel replication or interpolation.