ArcGIS Image Server 10.9.1 includes updates, enhancements, and bug fixes.
If you're upgrading from 10.8.1 or an earlier release to ArcGIS Image Server 10.9.1, you must reauthorize your software. You can obtain a new license file from My Esri.
For updates in Image Server 10.9, see here.
New at 10.9.1
See what's new and improved in ArcGIS Image Server 10.9.1.
Raster analysis
The updates to raster analysis capabilities are listed below.
Portal analysis tools in ArcGIS Pro
When you are signed in from ArcGIS Pro to an ArcGIS Enterprise portal that has ArcGIS Image Server configured for Raster Analysis, you can access the Raster Analysis tools. For example, many raster analysis tools that output a raster now support the Pyramid environment.
Suitability Modeler in ArcGIS Pro
When you are signed in from ArcGIS Pro to an ArcGIS Enterprise portal that has ArcGIS Image Server configured for Raster Analysis, you can access the Suitability Modeler using ArcGIS Pro as a client. You can work seamlessly with local raster datasets and web imagery layers running locally or on ArcGIS servers.
Raster functions in ArcGIS Pro
When you are signed in from ArcGIS Pro to an ArcGIS Enterprise portal that has ArcGIS Image Server configured for Raster Analysis, you can access new raster functions as well as additional capabilities that have been added to some existing functions.
The following new raster functions have been added:
- Dimensional Moving Statistics—Calculates statistics over a moving window on multidimensional data along a specified dimension
- Interpolate Raster By Dimension—Interpolates a multidimensional raster at specified dimension values, using adjacent slices
For existing functions, the following updates were made:
- Two new parameters were added to the Aggregate Multidimensional Raster function: the Percentile option was added to the Operation parameter, which enables the new Percentile Value and the Percentile interpolation type parameters.
- Cell Statistics—A new statistics operation of Percentile, which enables the new Percentile Value parameter for setting the percentile value to use. When the statistics operation is set to Median or Percentile, the new parameter Percentile Interpolation Type is enabled, with which you can determine the type of percentile interpolation to be used.
- Two new parameters were added the Compute Change function: the Field Name For Class Names In From Raster and Field Name For Class Names In To Raster parameters used to update field names accounting for class transitions. Additionally, the Use Color Method parameter was changed to Define Transition Colors.
- The Composite Bands function was enhanced with a new Cellsize Type parameter.
Raster analysis in ArcGIS REST API
The following raster analysis tasks are new:
- TrainRandomTreesRegressionModel—Models the relationship between explanatory variables (independent variables) and a target dataset (dependent variable).
- PredictUsingRegressionModel—Predicts data values using the output from the TrainRandomTreesRegressionModel tool.
- LocateRegions—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.
- SummarizeCategoricalRaster—Generates a table containing the pixel count for each class, in each slice of an input categorical raster.
The following raster analysis tasks have been enhanced:
- Three new parameters were added to the AggregateMultidimensionalRaster task: the Percentile option was added to the aggregation_method parameter, which enables the new percentile_value and the percentile_interpolation_type parameters. The dimensionless parameter specifies whether the layer will have dimension values.
- Two new parameters were added to the ComputeChangeRaster task: the fromClassnameField and toClassnameField parameters used to update field names accounting for class transitions. Additionally, the Use Color Method parameter was changed to Define Transition Colors.
Linux
A Linux server can now be configured as a Raster Analytics server.
Deep Learning
Native Linux Python configurations for Raster Analytics Deep Learning workflows are supported.
- Ampere GPU and CUDA 11 are supported.
ArcGIS Ortho Maker
Enhancements to ArcGIS Ortho Maker include the following:
- You can now create products in batch mode, after project creation.
- Pixel values can be compared between different flight lines, either in swipe mode or side-by-side mode.
- Advanced camera calibration options are now supported.
- Download products from the map viewer in GeoTiff format.
- Support has been added for batch processing when creating a project or adding a new flight to a project.
New at 10.9
See what's new in ArcGIS Image Server 10.9
Image service
Image services now support two new parameters that allow you to manage your imagery data, Allow Copy and Allow Analysis. Additional updates to image service functionality are listed below.
Image service operations in ArcGIS REST API
New image service operations are listed below:
- computeMultidimensionalInfo constructs a multidimensional info object for an image service of a mosaic dataset based on its catalog table.
- DownloadRaster is used to download an image or partial image at a designated resolution.
The identify operation now supports multidimensional raster data and can be configured with the new processAsMultidimensional parameter.
Create imagery content in your portal
The Create Imagery Layer wizard in your portal now includes a JPEG compression option for supported data types. The option is available on the Configure properties settings page. Additionally, the vertical coordinate system definition is now supported if you create a hosted imagery layer from raster elevation data. This option is available on the Configure properties settings page on the Spatial Reference tab.
ArcGIS Ortho Maker
Enhancements to ArcGIS Ortho Maker include the following:
- Support for multiple flights in one project.
- Same location for multiple flights collected on different dates.
- Each flight is processed separately, using its own parameter settings for producing products.
- A tool is provided to visually compare products from different flights.
- Manage ground control points at the project level for individual flights.
- Support for batch processing when creating a project or adding a new flight to a project.
Raster analysis
The updates to raster analysis capabilities are listed below.
Analysis tools in Map Viewer Classic
The raster analysis tools in Map Viewer Classic described below have been added or updated.
Analyze Patterns toolset
The Calculate Density tool has a new optional parameter, Choose barrier features, that alters the influence of a feature when calculating the density.
Multidimensional Analysis toolset
The Generate Trend Raster tool has two new options for the type of line to fit along a dimension: Mann-Kendall and Seasonal-Kendall.
Summarize Data toolset
The new Zonal Statistics as Table tool summarizes the values of a raster within the zones of another dataset and reports the results as a table.
Use Proximity toolset
The following five new tools have been added:
- Distance Accumulation—Calculates the accumulated distance for each cell to sources, allowing for straight-line distance, cost distance, and true surface distance, as well as vertical and horizontal factors
- Distance Allocation—Calculates distance allocation for each cell to the provided sources based on straight-line distance, cost distance, and true surface distance, as well as vertical and horizontal factors
- Optimal path As Line—Calculates the optimal path from destinations to sources as a line
- Optimal Path As Raster—Calculates the optimal path from destinations to sources as a raster
- Optimal Region Connections—Calculates the optimal connectivity network between two or more input regions
The Determine Optimum Travel Cost Network tool has been enhanced. The Calculate Distance tool has also been enhanced when the Distance Method parameter is set to Planar.
Raster functions in Map Viewer Classic
You can access raster functions in the Raster Analysis pane in Map Viewer Classic. The following new raster functions have been added:
- Boundary Clean smooths the boundaries between zones in raster data.
- LandTrendr Analysis evaluates changes in pixel values over time using the Landsat-based detection of trends in disturbance and recovery (LandTrendr) methods and generates a change analysis raster containing model results.
For existing functions, the following updates were made:
- Detect Change Using Change Analysis—Supports the output from the new LandTrendr Analysis raster function.
- Distance Accumulation and Distance Allocation—Performance enhancements when the Distance Method parameter is set to Planar.
- Trend Type parameter in Generate Trend—Supports analyzing the presence of trends using the Mann-Kendall and Seasonal Kendall methods.
- Kernel Density—A new Input Barriers parameter has been added that can be used to alter the influence of a feature when calculating the density.
- Least Cost Path—Performance enhancements.
- Merge Rasters—Supports merging multidimensional rasters across variables and dimensions.
- Zonal Statistics—Supports the Median and Percentile statistics types for floating point value rasters. A new parameter, Percentile Interpolation type has been added that defines the method of percentile interpolation to be used when the number of values from the input raster to be calculated are even. The function now also supports multidimensional processing for inputs with different dimensions.
Portal analysis tools in ArcGIS Pro
When you are signed in from ArcGIS Pro to an ArcGIS Enterprise portal that has an ArcGIS Image Server configured for Raster Analysis, you can access the Raster Analysis tools. The updates described below have been made to existing tools.
Analyze Patterns toolset
The Calculate Density tool has a new optional parameter, Choose barrier features, that alters the influence of a feature when calculating the density.
Summarize Data toolset
The Summarize Raster Within and Zonal Statistics As Table tools now support the Median and Percentile statistics types for floating point value rasters. A new parameter, Percentile Interpolation Type, has been added to these tools to determine the type of percentile interpolation when the number of values from the input value raster to be calculated are even. These tools now also support multidimensional processing for inputs with different dimensions.
Use Proximity toolset
- Distance Accumulation, Distance Allocation, and Optimal Region Connections—Performance enhancements when the Distance Method parameter is set to Planar.
- Calculate Travel Cost, Determine Optimum Travel Cost Network, and Determine Travel Cost Path As Polyline—Performance enhancements.
- Optimal Path As Line—A new Create network paths parameter has been added that specifies whether complete, and possibly overlapping, paths from the destinations to the sources are calculated or if nonoverlapping network paths are created. The result from this new parameter can be used for networking.
Use Proximity (Legacy) toolset
The Calculate Distance tool has performance enhancements when the Distance Method parameter is set to Planar.
Raster functions in ArcGIS Pro
When you are signed in from ArcGIS Pro to an ArcGIS Enterprise portal that has an ArcGIS Image Server configured for Raster Analysis, you can access new raster functions as well as additional capabilities that have been added to some existing functions.
The following new raster functions have been added:
- Aggregate Multidimensional—Mathematically combines slices from multidimensional raster data over given dimension values
- Boundary Clean—Smooths the boundaries between zones in raster data
- LandTrendr Analysis—Evaluates changes in pixel values over time using the Landsat-based detection of trends in disturbance and recovery (LandTrendr) methods and generates a change analysis raster containing model results
For existing functions, the following updates were made:
- Band Arithmetic—Includes six new index calculation options: Burn Area Index (BAI), Normalized Burn Ratio Index (NBR), Normalized Difference Built-up Index (NDBI), Normalized Difference Moisture Index (NDMI), Normalized Difference Snow Index (NDSI), and Normalized Difference Water Index (NDWI).
- Compute Change—Supports filtering the classes to generate for categorical change detection using the new Filter Method parameter.
- Detect Change Using Change Analysis—Supports the output from the new LandTrendr Analysis raster function.
- Distance Accumulation and Distance Allocation—Performance enhancements when the Distance Method parameter is set to Planar.
- Trend Type parameter in Generate Trend—Supports analyzing the presence of trends using the Mann-Kendall and Seasonal Kendall methods.
- Kernel Density—Includes a new Input Barriers parameter that can be used to alter the influence of a feature when calculating the density.
- Least Cost Path—Performance enhancements.
- Merge Rasters—Supports merging multidimensional rasters across variables and dimensions.
- Statistics—Supports new Majority and Minority fill methods.
- Zonal Statistics—Supports the Median and Percentile statistics types for floating point value rasters. A new parameter, Percentile Interpolation type, has been added that defines the method of percentile interpolation to be used when the number of values from the input raster to be calculated are even. The function now also supports multidimensional processing for inputs with different dimensions.
Raster analysis in ArcGIS REST API
The following raster analysis operations are new:
- ComputeAccuracyForObjectDetection
- ComputeChangeRaster
- MergeMultidimensionalRasters
- AnalyzeChangesUsingLandTrendr
The following raster analysis operations have been enhanced:
- The deep learning inferencing operations all support .dlpk file inputs from a raster store or raster store subfolder.
- The BuildMultidimensionalTranspose operation has a new parameter: deleteTranspose.
- The CalculateDensity operation has a new parameter, input_barrier_features, that can be used to alter the influence of a feature when calculating the density.
- The CalculateDistance, DistanceAccumulation, DistanceAllocation and OptimalRegionConnections operations have performance enhancements when the distance_method parameter is set to PLANAR.
- The CalculateTravelCost, DetermineOptimalTravelCostNetwork and DetermineTravelCostPathAsPolyline operations have performance enhancements.
- The DetectChangeUsingChangeAnalysisRaster operation supports a change analysis raster generated by the new AnalyzeChangesUsingLandTrendr task.
- The ExportTrainingDataForDeepLearning operation now supports the MultiLabeled_Tiles and Export_Tiles options for the output metadata format.
- The GenerateTrendRaster operation supports testing for the presence of trends using the Mann-Kendall and Seasonal Mann-Kendall methods, which are available in the trendLineType parameter.
- The TrainDeepLearningModel operation now automatically uses the GPU unless otherwise specified. Additionally, three new model types were added: YOLOV3, DEEPLAB, and FASTERRCNN. The backbone model option DARKNET53 is also supported.
- The SummarizeRasterWithin and ZonalStatisticsAsTable operations now support the MEDIAN and PERCENTILE statistics types for floating point value rasters. A new parameter, percentile_interpolation_type, has been added to these tasks to determine the type of percentile interpolation when the number of values from the input value raster to be calculated are even. These operations now also support multidimensional processing for inputs with different dimensions.
- Additional options were added to the context parameter for the following raster analysis operations: BuildMultidimensionalTranspose, DetectChangeUsingChangeAnalysisRaster, GenerateMultidimensionalAnomaly, GenerateTrendRaster, PredictUsingTrendRaster, and Sample.
- Additional options were added to the context parameter for the following ortho mapping operations: ComputeControlPoints, ComputeSeamlines, ComputeSensorModel, GenerateDEM, GenerateOrthomosaic, and MatchControlPoints.
Raster analysis in ArcGIS API for Python
New functionality has been added.
arcgis.raster.analytics module
The following functions have been enhanced:
- calculate_density—Includes a new parameter, input_barriers, that can be used to alter the influence of a feature while calculating density.
- optimal_path_as_line—Includes a new parameter, createNetworkPaths, that specifies whether complete, and possibly overlapping, paths from the destinations to the sources are calculated or if nonoverlapping network paths are created. The result from this new parameter can be used for networking.
- summarize_raster_within—Supports the Median and Percentile statistics types for floating point value rasters. A new parameter, percentile_interpolation_type, has been added to this function to determine the type of percentile interpolation when the number of values from the input value raster to be calculated are even. This function now also supports multidimensional processing for inputs with different dimensions.
Additionally, the optimum_travel_cost_network and determine_travel_costpath_as_polyline functions have performance enhancements. The optimal_region_connections function has performance enhancements when the distance_method parameter is set to Planar.
arcgis.raster.functions.gbl module
A new function, boundary_clean, smooths the boundary between zones in raster data.
Additionally, the following functions have been enhanced:
- calculate_distance, distance_accumulation and distance_allocation—Performance enhancements when the distance_method parameter is set to Planar.
- calculate_travel_cost and least_cost_path—Performance enhancements.
- kernel_density—Includes a new parameter, in_barriers, that can be used to alter the influence of a feature while calculating density.
- zonal_statistics—Supports the Median and Percentile statistics types for floating point value rasters. A new parameter, percentile_interpolation_type, has been added that determines the type of percentile interpolation when the number of values from the input value raster to be calculated are even. This function now also supports multidimensional processing for inputs with different dimensions.