- Operations:Add Image, Aggregate Multidimensional Raster, Analyze Changes Using CCDC, Analyze Changes Using LandTrendr, Build Multidimensional Transpose, Calculate Density, Calculate Distance, Calculate Travel Cost, Classify, Classify Objects Using Deep Learning, Classify Pixels Using Deep Learning, Compute Accuracy For Object Detection, Convert Feature to Raster, Convert Raster to Feature, Copy Raster, Cost Path as Polyline, Create Image Collection, Create Viewshed, Delete Image, Delete Image Collection, Derive Continuous Flow, Detect Change Using Change Analysis Raster, Detect Objects Using Deep Learning, Determine Optimum Travel Cost Network, Determine Travel Cost Path as Polyline, Determine Travel Cost Paths to Destinations, Distance Accumulation, Distance Allocation, Download Raster, Export Training Data for Deep Learning, Fill, Find Argument Statistics, Flow Accumulation, Flow Direction, Flow Distance, Generate Multidimensional Anomaly, Generate Raster, Generate Trend Raster, Install Deep Learning Model, Interpolate Points, Linear Spectral Unmixing, List Deep Learning Model, Locate Regions, Manage Multidimensional Raster, Merge Multidimensional Raster, Nibble, Optimal Path As Line, Optimal Path As Raster, Optimal Region Connections, Predict Using Regression Model, Predict Using Trend Raster, Publish Deep Learning Model, Query Deep Learning Model Info, Sample, Segment, Stream Link, Subset Multidimensional Raster, Summarize Categorical Raster, Summarize Raster Within, Surface Parameters, Train Classifier, Train Deep Learning Model, Train Random Trees Regression Model, Uninstall Deep Learning Model, Watershed, Zonal Statistics As Table
- Version Introduced:
The Raster Analysis service contains a number of tasks that you can access and use in your applications. These tasks are arranged below in categories of logical groupings, which do not affect how you access or use the tasks in any way.
Starting in ArcGIS Enterprise 10.6, an input image service can be secured. If your raster function requires a secured image service as an input, you must provide a token, and possibly a referrer, along with the URL so the analysis service can access it. A long-lived token can be obtained from the token server. For more details, see Acquire ArcGIS tokens.
Tasks that analyze patterns
The tasks that analyze patterns are described in the following table:
The CalculateDensity task creates a density layer from point features by spreading known quantities of some phenomenon (represented as attributes of the points) across the raster. The result is a layer of areas classified from least dense to most dense.
The ComputeChangeRaster task is used to evaluate the difference between two input rasters, and generates a change raster output service. This tool supports continuous raster data and categorical raster data. When the raster inputs are two categorical rasters, the output change raster includes an attribute table containing all the permutations of the from and to classes as well as the pixel counts for all changed and unchanged classes.
The InterpolatePoints task allows you to predict values at new locations based on measurements from a collection of points. The tool takes point data with values at each point and returns a raster of predicted values.
Tasks that analyze terrain
The tasks that analyze terrain are described in the following table:
The CreateViewshed task uses an elevation surface and observer locations to identify areas where the observers can see the observed objects and the observed objects can see the observers.
The SurfaceParameters task determines parameters of a surface raster such as various types of curvatures, slope, and aspect.
Tasks that classify data
The tasks that perform image classification are described in the following table:
The Classify task creates categories of pixels based on the input raster and the classifier definition JSON that was generated from the TrainClassifier service.
The LinearSpectralUnmixing task performs subpixel classification and calculates the fractional abundance of end members for individual pixels.
The Segment task groups adjacent pixels that have similar spectral and spatial characteristics into segments.
The TrainClassifier task is a service to train image classifiers in a deep learning model and return an .ecs file in JSON format. The .ecs file is used in the Classify task.
Models the relationship between explanatory variables (independent variables) and a target dataset (dependent variable).
Tasks that perform deep learning analysis
The tasks that perform deep learning are described in the following table:
The ClassifyObjectsUsingDeepLearning task is used to classify objects based on overlaid imagery data using the designated deep learning model and generate a feature service with a new assigned label for each object.
The ClassifyPixelsUsingDeepLearning operation can be used to classify pixels in the imagery data using the designated deep learning model and generate an image service for the classified raster.
The ComputeAccuracyForObjectDetection task is used to calculate the accuracy of a deep learning model by comparing the detected objects from the Detect Object Using Deep Learning tool to ground truth data.
The DetectObjectsUsingDeepLearning operation can be used to detect objects from the imagery data using the designated deep learning model and generate a feature service for the detected objects.
The ExportTrainingDataforDeepLearning service generates training sample image chips from the input imagery data with labeled vector data or classified images. The output of this service tool is the data store string where the output image chips, labels, and metadata files will be stored.
The InstallDeepLearningModel operation is used to install the uploaded deep learning model package (.dlpk) from portal to the raster analysis image server. The upload model package will be unpacked and saved to the server configuration store.
The ListDeepLearningModels operation is used to list all the installed deep learning models on the raster analysis image server.
The PublishDeepLearningModel operation publishes a model package of a deep learning model (.dlpk) containing the files and data required to run deep learning inferencing tools for object detection or image classification to your portal as a DLPK item.
The QueryDeepLearningModelInfo operation is used to extract the deep learning model specific settings from the model package item or model definition file.
The TrainDeepLearningModel task is used to train a deep learning model using the output from the ExportTrainingDataforDeepLearning operation. It generates the deep learning model package (*.dlpk) and adds it to an enterprise portal. You can also use this task to write the deep learning model package to a file share data store location.
The UninstallDeepLearningModel operation is used to uninstall the uploaded deep learning model package (.dlpk) from the raster analysis image server. It will delete the named deep learning model from the image server's configuration store but not the portal item.
Tasks that generalize raster data
The task that generalizes raster data is described in the following table:
The Nibble task replaces the input cells corresponding to a mask with the values of the nearest neighbors.
Tasks that perform hydrology analysis
The tasks that perform hydrology analysis are described in the following table:
The DeriveContinuousFlow task generates a raster of accumulated flow into each cell from an input surface raster with no prior sink or depression filling required.
The Fill task fills sinks in a surface raster to remove small imperfections in the data.
The FlowAccumulation task creates a raster of accumulated flow into each cell. A weight factor can optionally be applied.
The FlowDirection task creates a raster of flow direction from each cell to its steepest downslope neighbor.
The FlowDistance task computes the downslope horizontal or vertical distance to cells in a stream or river into which they flow. A flow direction raster can optionally be applied. In case of multiple flow paths, minimum, weighted mean, or maximum flow distance can be computed.
The StreamLink task assigns unique values to sections of a raster linear network between intersections.
The Watershed task determines the contributing area above a set of cells in a raster.
Tasks that manage data
The tasks that manage data are described in the following table:
The AddImage operation allows you to add new images to an existing image collection.
The ConvertFeatureToRaster task converts a point, line, or polygon feature dataset to a raster.
The ConvertRasterFunctionTemplate task converts a raster function template between JSON and XML formats. ConvertRasterFunctionTemplate takes a raster function template in any of these two formats as input to convert the original source into the other specified format.
The ConvertRasterToFeature task converts a raster to a point, line, or polygon feature dataset.
The CopyRaster task takes single raster layer input and generates the output image using parallel processing.
The input raster dataset can be clipped, resampled, and reprojected based on the setting.
The CreateImageCollection task takes multiple image items as input, creates a image collection in a registered data store, and publishes it as an image service. The input raster dataset can be clipped, resampled, and reprojected based on the setting. The image upload can also be run in parallel.
The DeleteImage task allows you to remove existing images from the image collection. The service will only delete the raster item in the mosaic dataset and will not remove the source image.
The DeleteImageCollection task deletes the image collection image service, that is, the portal-hosted image layer item. It will not delete the source images that the image collection references.
The DownloadRaster task is used to download an image or partial image at a designated resolution. The input image service must be configured to allow pixel data download.
The GenerateRaster task is a service that allows you to execute raster analysis on a distributed server deployment. The analysis can be specified either with a predefined server raster function keyword or by giving a JSON object representation of a raster function chain.
Tasks that process multidimensional raster data
The tasks that analyze or manage multidimensional raster data are described in the following table:
The AggregateMultidimensionalRaster task can be used to generate a .CRF multidimensional raster dataset and image service by aggregating existing multidimensional dataset variables along a dimension.
The AnalyzeChangesUsingCCDC task evaluates changes in pixel values over time using the CCDC algorithm, and generates a multidimensional raster containing the model results.
The AnalyzeChangesUsingLandTrendr task evaluates changes in pixel values over time using the Landsat based detection of trends in disturbance and recovery (LandTrendr) method and generates a change analysis raster containing the model results.
The BuildMultidimensionalTranspose task transposes a multidimensional raster dataset, which divides the multidimensional data along each dimension to optimize performance when accessing pixel values across all slices.
The DetectChangeUsingChangeAnalysisRaster task generates a raster containing pixel change information using the output change analysis raster from the AnalyzeChangesUsingCCDC task or the AnalyzeChangesUsingLandTrendr task.
The FindArgumentStatistics task is used to extract the dimension value or band index at which a given statistic is attained for each pixel in a multidimensional or multiband raster.
The GenerateMultidimensionalAnomaly task is used to compute the anomaly for each slice in a multidimensional raster to generate a multidimensional raster. An anomaly is the deviation of an observation from its standard or mean value.
The GenerateTrendRaster task allows you estimate the trend for each pixel along a dimension for one or more variables in a multidimensional raster.
The ManageMultidimensionalRaster task edits a multidimensional raster by adding or deleting variables or dimensions.
The MergeMultidimensionalRasters task combines multiple multidimensional raster datasets spatially or across variables and dimensions.
Predicts data values using the output from the TrainRandomTreesRegressionModel tool.
The PredictUsingTrendRaster task is used to compute a forecasted multidimensional raster using the output trend raster from the Generate Trend Raster tool.
The SubsetMultidimensionalRaster task creates a subset of a multidimensional raster by slicing data along defined variables and dimensions.
Tasks that overlays data
The task that overlays data is described in the following table:
The LocateRegions task identifies the best regions, or groups of contiguous cells, from an input utility (suitability) raster that satisfy a specified evaluation criterion and that meet identified shape, size, number, and interregion distance constraints.
Tasks that summarize data
The tasks that summarize data are described in the following table:
The Sample task creates a table of cell values from a raster, or set of rasters, for defined locations. The locations are defined by raster cells, polygon features, polyline features, or by a set of points.
Generates a table containing the pixel count for each class, in each slice of an input categorical raster.
The SummarizeRasterWithin task summarizes the cells of a raster within the boundaries of zones defined by another dataset.
The ZonalStatisticsAsTable task summarizes the cells of a raster within the boundaries of zones defined by another dataset.
Tasks that use proximity for performing analysis
The tasks that use proximity for performing analysis are described in the following table:
The DistanceAccumulation task calculates accumulated distance for each cell to sources, allowing for straight-line distance, cost distance, true surface distance, as well as vertical and horizontal cost factors.
The DistanceAllocation task calculates distance allocation for each cell to the provided sources based on straight-line distance, cost distance, true surface distance, as well as vertical and horizontal cost factors.
The OptimalPathAsLine task calculates the optimal path from a source to a destination as a feature.
The OptimalPathAsRaster task calculates the optimal path from a source to a destination as a raster.
The OptimalRegionConnections task calculates the optimal connectivity network between two or more input regions.
Legacy tasks that use proximity for performing analysis
Legacy tasks that use proximity for performing analysis are described in the following table:
The CalculateDistance task calculates the Euclidean distance, direction, and allocation from a single source or set of sources.
The CalculateTravelCost task calculates the cost distance from a single source or set of sources, while accounting for surface distance and horizontal and vertical cost factors.
The CostPathAsPolyline task calculates the least-cost path from a source to a destination.
The DetermineOptimumTravelCostNetwork task calculates the optimum cost network from a set of input regions.
The DetermineTravelCostPathAsPolyline task calculates the least-cost path between sources and destinations.
The DetermineTravelCostPathsToDestinations task calculates specific paths between known sources and known destinations.