The June 2017 update of ArcGIS Online includes some useful capabilities for displaying imagery served by your image services. These capabilities give you greater control for visualizing the information contained in your image services. When we talk about rendering, we’re not talking about making soap out of fat. Here at Esri, rendering is the process of displaying your data. How an image service is rendered depends on what type of data it contains and what you want to show.
Once you search for and add a layer, and your image is displayed in Map Viewer, click the More Options icon then Display to open the Image Display pane.
You see a new category named Image Enhancement. This is where the real fun begins.
The Symbology Type options include Unique Values, Stretch and Classify. Unique Values and Classify renderers work with single-band image services, while the Stretch renderer works on both single and multiple band images.
Unique Values Renderer
Unique values symbolize each value in the raster layer individually and are supported on single band layers with Raster Attribute table. The symbology can be based on one of more attribute fields in the dataset. The colors are read from the Raster Attribute table and if they are not available the renderer assigns a color to each value in your dataset. This symbology type is often used with single band thematic data, such as land cover, because of its limited number of categories. It can also be used with continuous data if you choose a color ramp that is a gradient.
- Use the Field drop-down to select the field you want to map. The field is displayed in the table.
- Click the Color Ramp drop-down and click on a color scheme. If your image service already has a color ramp, such as the NLCD service in this example, it is displayed by default.
- The colors in the Symbol column and Labels can be edited as required.
- Click Apply to display the rendering in the layer.
The stretch parameters improve the appearance of your image by adjusting the image histogram controlling brightness and contrast enhancements. Either single or multiple band images can be stretched. For multiple band images, the stretch is applied to the band combination previously chosen in the RGB Composite options. The stretch options enhance various ground features in your imagery to optimize information content.
- Click the Stretch Type drop-down arrow and choose the stretch type to use. The following contrast enhancements determine the range of values that are displayed.
- None – No additional image enhancement will be performed
- Minimum and Maximum – Displays the entire range of values in your image. Additional changes can be made by editing the values in the Min-Max grid (available only when Dynamic range adjustment is turned off.)
- Standard Deviation – Display values between a specified number of standard deviations
- Percent Clip – Set a range of values to display. Use the two text boxes to edit the top and bottom percentages.
- If the Stretch type is set to an option other than None, the following additional image enhancement options will be available
- Dynamic range adjustment – Performs one of the selected stretches, but limits the range of values to what is currently in the display window. This option is always turned on if the imagery layer does not have global statistics.
- Gamma – Stretches the middle values in an image but keeps the extreme high and low values constant.
- For single-band layers, you can optionally choose a new color scheme from the Color Ramp drop-down menu after applying a stretch method on the layer.
- Click Apply to display the rendering in the layer.
Here’s a WorldView-2 natural color image of Charlotte, NC, using the default no stretch:
And here is the same imagery layer with the top 2% and bottom 20% of the histogram omitted:
Classify symbology is supported by single band layers. It allows you to group pixels together in a specified number of classes. The following are the different settings available with the Classify symbology.
- Field – Represents the values of the data.
- Method – Refers to how the break points are calculated.
- Defined Interval – You specify an interval to divide the range of pixel values and the number of classes will be automatically calculated.
- Equal Interval – The range of pixel values are divided into equally sized classes where you specify the number of classes.
- Natural Breaks – The class breaks are determined statistically by finding adjacent feature pairs between which there is a relatively large difference in data value.
- Quantile – Each class contains equal number of pixels.
- Classes – Sets the number of groups.
- Color Ramp – Allows you to choose the color ramp for displaying the data.
Classify symbology works with single band layers that have either a Raster Attribute Table or Histogram values. If a histogram is absent, it is generated when you select the symbology type.
Here’s the classified map of Charlotte, specifying 15 classes and using the Natural Breaks method for determining class breaks:
These new Map Viewer image rendering capabilities are similar to what you are used to in ArcMap and ArcGIS Pro. Since this release, Scene Viewer also supports imagery layers, however we are still working on bringing the new Map Viewer image rendering capabilities into Scene Viewer. Check out these new imagery capabilities in ArcGIS Online and see how they can enhance the stories behind your data.
Please leave us comments below for any future enhancements you’d like to see. And check back in a few months; we have a lot of other cool stuff planned for imagery in upcoming releases.