Tag: statistics

Server Object Interceptor (SOI) to Visualize Features with On The Fly Aggregated Values from Time-Series or Historical Observation Data

aggregated layer no UI

A sample SOI (Server Object Interceptor) is made available that helps visualize on the fly aggregated results by overcoming challenges presented in an earlier blog post. A map service, powered by this SOI, can be used in the ArcGIS Online Map Viewer … Continue reading

Posted in Apps, ArcGIS Enterprise, ArcGIS Online, Developer, Hydro, Mapping, Portal for ArcGIS, Services, Spatial Statistics, Water Utilities, Web | Tagged , , , , , , , , , , , , , , , , , , | Leave a comment

My source mosaic dataset items already have statistics, why would I need to re-calculate them?

color balanced

The main reason you will need statistics for each raster dataset is because you want to color balance your mosaic dataset. The mosaic dataset items may have been radiometrically altered through a raster function, and therefore needs the most up … Continue reading

Posted in Imagery | Tagged , , | Leave a comment

Do I need to calculate statistics on a mosaic dataset after a raster function is applied?

Calculating statistics are not always required after you apply a function.  However, there are times when it may be helpful or even required.

When are statistics required?
Color correction requires up-to-date statistics.  Therefore if any functions have been applied to your mosaic dataset, since statistics were last calculated, you will need to calculate them again.

When are statistics a good idea?
Often times when you apply a function, it can change the (virtual) pixel values. If you want to keep your mosaic dataset using the most accurate statistics and rendering, then you will need to calculate statistics again.

For instance, when you use the Stretch function to map your 8-bit data to a 16-bit pixel depth, your images will probably show up black, unless you calculate statistics.

Written by: Simon Woo and Jie Zhang

Posted in Imagery | Tagged , , , , , | Leave a comment

A Model to Filter Images for Build Pyramids and Statistics Tool

In ArcGIS 10.0, we introduced a new raster geoprocessing tool Build Pyramids and Statistics. You can use this tool to build pyramids and calculate statistics for all the raster datasets in a workspace, and even in a raster catalog or a mosaic dataset.

This tool provides a convenient solution to build multiple raster datasets in single step. However it works on all raster datasets in the workspace. Although the tool does not contain a filter option to allow you select a smaller collection of raster among all, we can design a simple model to do this.

This is what the model could look like:

Continue reading

Posted in Analysis & Geoprocessing, Imagery | Tagged , , , , , , , | 2 Comments

Raster XML file contains a histogram in ArcGIS 10

At 10.0, ArcGIS began storing the histogram for a raster
dataset when it generates the statistics.  This allows the application to
provide more capabilities, such as adding stretches, like Percent Clip.  If you export the statistics file you can take a look all
the information.

Previously to 10.0 the exported XML file would look like…
<?xml version=”1.0″?>
<RasterStatistics xml:lang=”en”>
<Band Name=”dem30″>

At 10.0 the exported XML file will contain the histogram…
<?xml version=”1.0″?>
<RasterStatistics xml:lang=”en”>
<Band Name=”dem30″>
. . .


Submitted by: Melanie Harlow

Posted in Imagery | Tagged , , , , , , | Leave a comment

Does your raster data display as a black rectangle on your screen?

One of the most common issues that seem to plague our users, is their raster data displaying as a black rectangle.  This is especially problematic for data that is 16-bit or 32-bit.

Here is an over simplified flowchart to help you take the first steps in solving your black rectangle problem.


Hopefully  this helped you solve your issue, or at least steered you in the right direction. ArcGIS can only render your data based on the pixel values of your dataset. If all your pixel values are zero (or near zero) then it will render it as such. 


Posted in Imagery | Tagged , , , , , , , | Leave a comment

Advanced labeling for the stretch renderer


Do you like the visualization of the stretched renderer, but wish you could label more than three values? The Labeling button on the stretched renderer allows you to create breakpoints, choose a color for each breakpoint, and chooe a label for each breakpoint. It is like a hybrid between the stretched renderer and classified renderer. Continue reading

Posted in Imagery, Mapping | Tagged , , , , , | 26 Comments

Are the statistics for the classified renderer accurate?

When using the classified renderer, some people notice that the statistics in the Classification dialog are not exactly the same as the statistic of the Raster Dataset Properties. Most often, this is because your classified renderer is using an approximation for its statistics. Even though the statistics are estimated, most times your rendered output is quite similar.

How do I turn on accurate statistics for my classified renderer?
In order to change the classified renderer to use accurate statistics, you will need to set a registry setting.  KB article 35443 shows you the steps to follow, to set your registry to use accurate statistics when using the classified renderer.

Why wouldn’t I always want this on?
When the accurate statistics is on, the rendering performance slows down. And usually the estimated statistics for the classified renderer is sufficient enough for a similar classification.

When using the accurate statistics, sometimes a cryptic warning message would appear.

If this message appears, then you will need to follow the steps in KB Doc 35053.

You can try both accurate and estimated classified renderer statistics, and see which one is a better workflow for you.  Most users have told us they prefer the faster estimate (which is the default). Others require absolute accuracy when classifying.


Submitted by: Simon Woo


Posted in Analysis & Geoprocessing, Imagery | Tagged , , , | 2 Comments