Compute Control Points (Data Management)


Creates the control points between the mosaic dataset and the reference image. The control points can then be used in conjunction with tie points to compute the adjustments for the mosaic dataset.


  • For accurate control point results, use the High similarity option for the Similarity parameter.

  • The control points can be combined with tie points, using the Append Control Points tool.

  • The control points and tie points are then used within the Compute Block Adjustment tool.

  • If you have a mosaic dataset with many items, use caution when specifying the Output Image Features parameter, since your result may take a long time to process.


ComputeControlPoints(in_mosaic_dataset, in_reference_images, out_control_points, {similarity}, {out_image_feature_points}, density, distribution, area_of_interest, {location_accuracy})
ParameterExplanationData Type

The input mosaic dataset that will be used to create control points.

Mosaic Dataset; Mosaic Layer

The reference images that will be used to create control points for your mosaic dataset. If you have multiple images, create a mosaic dataset from the images and use the mosaic dataset as the reference.

Raster Layer; Raster Dataset; Image Service; MapServer; WMS Map; Mosaic Layer; Internet Tiled Layer; Map Server Layer

The output control point table. This table will contain the control points that were created.

Feature Class

Specifies the similarity level for matching tie points.

  • LOWThe similarity criteria for the two matching points will be low. This option will produce the most matching points, but some of the matches may have a higher level of error.
  • MEDIUMThe similarity criteria for the matching points will be medium.
  • HIGHThe similarity criteria for the matching points will be high. This option will produce the least number of matching points, but each matching will have a lower level of error.

The output image feature points table. This will be saved as a polygon feature class. This output can be quite large.

Feature Class

The number of tie points to be created.

  • LOWSet the density of points to be low. This will create the fewest number of tie points.
  • MEDIUMSet the density of points to be medium. This will create a moderate number of points.
  • HIGHSet the density of points to be high. This will create the highest number of points.

Specifies whether the points will have regular or random distribution.

  • RANDOMPoints are generated randomly. Randomly generated points are better for overlapping areas with irregular shapes.
  • REGULARPoints are generated based on a fixed pattern. Points based on a fixed pattern use the point density to determine how frequently to create points.

Limit the area in which tie points are generated to only this polygon feature class.

Feature Layer

Specifies the keyword that describes the accuracy of the imagery.

  • LOWImages have a large shift and a large rotation (> 5 degrees).The SIFT algorithm will be used in the point-matching computation.
  • MEDIUMImages have a medium shift and a small rotation (<5 degrees).The Harris algorithm will be used in the point-matching computation.
  • HIGHImages have a small shift and a small rotation.The Harris algorithm will be used in the point-matching computation.

Code sample

ComputeControlPoints example 1 (Python window)

This is a Python sample for the ComputeControlPoints tool.

import arcpy
     "c:/block/BD.gdb/redQB_tiePoints", "HIGH",
     "c:/block/BD.gdb/redQB_mask", "c:/block/BD.gdb/redQB_imgFeatures")
ComputeControlPoints example 2 (stand-alone script)

This is a stand-alone script sample for the ComputeControlPoints tool.

#compute control points

import arcpy
arcpy.env.workspace = "c:/workspace"

#compute control points using a mask 
mdName = "BD.gdb/redlandsQB"
in_mask = "BD.gdb/redlandsQB_mask"
out_controlPoint = "BD.gdb/redlandsQB_tiePoints"
out_imageFeature = "BD.gdb/redlandsQB_imageFeatures"

arcpy.ComputeControlPoints_management(mdName, out_controlPoint, 
     "HIGH", in_mask, out_imageFeature)

Licensing information

  • Basic: No
  • Standard: Yes
  • Advanced: Yes

Related topics