The Detect Target Using Spectra tool computes a matching score for each pixel in an input raster based on how well the pixel's spectrum matches the target spectra defined by an input spectral signature file.
The output is a hosted imagery layer.
Examples
Identify the location of clay minerals in a hyperspectral satellite image using the spectrum of the mineral as input.
Usage notes
Detect Target Using Spectra includes configurations for the input layer, detection settings, and the result layer.
Input layer
The Input layer group includes the following parameters:
- Input raster specifies the multiband raster containing the target pixels to be detected.
- Input spectra file type indicates the file type of the input spectral signature file. The following options are available:
- Spectra file (default)—The spectral signature file is an Esri spectral library file (.esl).
- Feature layer—The spectral signature file is a point feature layer, which you can create in ArcGIS Pro using Training Samples Manager.
- Input spectra file specifies the spectral signature file that contains the target spectra.
Detection settings
The Detection settings group includes the following parameters:
- Detection method specifies the method used to compute the matching score between the input raster and the spectral signature file. All scores are normalized from zero to one. A higher score indicates a better match between a pixel's spectrum in the input raster and the target spectrum in the spectral signature file. The following options are available:
- SAM (default)—Use the Spectral Angle Mapper method to compute matching scores. This method measures the angle between the target spectrum and the pixel spectrum. It is not sensitive to changes in illumination.
- SID—Use the Spectral Information Divergence method to compute matching scores. This method measures the divergence between the probability distributions of the target spectrum divided by the pixel spectrum. This option is useful to identify targets in pixels with mixed spectra.
- SID-SAM—Use a combination of the Spectral Information Divergence and Spectral Angle Mapper methods to compute matching scores. This method can discriminate targets in pixels with mixed spectra better than the SID and SAM options individually.
- NS3—Use the Normalized Spectral Similarity Score method to compute matching scores. This method uses the Euclidian distance and the SAM score to compute the matching score. This method can discriminate targets efficiently but requires extensive reference data for accuracy.
- CEM—Use the Constrained Energy Minimization method to compute matching scores. This method uses a Finite Impulse Response (FIR) filter, which has a greater response to a pixel spectrum when it is more similar to the target spectrum. This option is useful when the background noise is minimal.
- ACE—Use the Adaptive Cosine Estimator method to compute matching scores. This method estimates the similarity between a target vector and a pixel's spectrum after normalizing for background noise. This option is useful when the input raster has significant background noise.
- MF—Use the Matched Filter method to compute matching scores. This method uses a FIR filter, which has a greater response to a mean-subtracted pixel spectrum when it is more similar to the mean-subtracted target spectrum. This option is useful when the background noise is additive white Gaussian noise.
Remove Continuum indicates whether to normalize the spectra, which can be done using an image or reference data. By default, this parameter is unchecked, and the spectra are not normalized. If checked, the spectra are normalized.
To create the continuum, the tool uses straight-line segments to connect local maxima in the spectrum and fits a convex hull over the top of the spectrum. To remove the continuum, the continuum is divided into the spectrum of each pixel in the input raster and the target spectra in the input spectral signature file. This removal allows you to compare the absorption features in the input raster and input spectral signature file from a common baseline.
Result layer
The Result layer group includes the following parameters:
- Output score raster specifies the name of the layer containing the matching score raster that is created and displayed. The name must be unique. If a layer with the same name already exists in your organization, the tool will fail, and you will be prompted to use a different name.
- Output layer type specifies the type of raster output that will be created. The output can be either a tiled imagery layer or a dynamic imagery layer.
- Save in folder specifies the name of a folder in My content where the result will be saved.
Limitations
The following limitations apply to the tool:
- If the input spectral signature file is a point feature layer, the number of bands in the input raster and input spectral signature file must match.
- If the input spectral signature file is an .esl file, and the number of bands in the file do not match the number of bands in the input raster, the spectral data in the input spectral signature file is resampled using a band averaging method. Alternatively, you can resample the spectra using the Resample Library Spectra tool.
- If the input spectral signature file is from a spectral library in which reflectance values range from zero to one, such as the USGS spectral library, the input raster must also contain values between zero and one.
- If the pixel value range of the input raster is not within zero to one, such as in 8-bit or 16-bit imagery, it is recommended that you turn on Remove Continuum to normalize the spectra. Alternatively, you can collect spectral signatures of the target material using the Spectral Viewer tool in ArcGIS Pro, which allows you to save spectra in a range defined by an input raster.
Environments
Analysis environment settings are additional parameters that affect a tool's results. You can access the tool's analysis environment settings from the Environment settings parameter group.
This tool honors the following analysis environments:
- Output coordinate system
- Geographic transformations
- Processing extent
Note:
The default processing extent for raster analysis tools and raster functions is Display extent. This default is different from other types of analysis.
- Snap raster
- Cell size
- Resampling method
Credits
This tool consumes credits.
Use Estimate credits to calculate the number of credits that will be required to run the tool. For more information, see Understand credits for spatial analysis.
Outputs
The output includes one imagery layer that represents the matching scores between the input raster and the spectral signature file. If the input spectral signature file contains multiple spectral signatures, the output is multiband with one band per signature.
Usage requirements
This tool requires the following user type and configurations:
- Creator, Professional, or Professional Plus user type
- Publisher, Facilitator, or Administrator role, or an equivalent custom role
Resources
Use the following resources to learn more:
- Detect Target Using Spectra in ArcGIS API for Python
- Detect Target Using Spectra in ArcGIS Pro