The RGB Composite renderer uses the same methods as the Stretched renderer but allows you to combine bands as red, green, blue composites. When viewing color aerial photography, you are often viewing a three-band raster dataset, and this renderer is applied by default. You will also want to use this renderer to display different combinations of bands when working with multiband raster datasets, such as satellite or aerial imagery.
You can make modifications to this renderer by accessing the layer's properties or using the Image Analysis window.
The RGB Composite renderer uses statistics to apply the stretch. When you display a raster dataset with statistics, they will be used. If the raster datasets do not have statistics, then they will be sampled, but only for the purpose of displaying. To render the raster dataset, by taking into account all cell values, the raster dataset must have statistics.
Learn about calculating statistics
All renderers have an Import 
button. Click this button to import symbology from a raster layer file (*.lyr) or a raster layer that is currently in the table of contents. The current renderer must match the renderer of the imported layer.
- In the table of contents, right-click the multiband raster layer you want to draw as an RGB composite and click Properties.
- Click the Symbology tab.
- Click RGB Composite.
- Click the Band drop-down arrow next to each color channel and click the band you want to display for that color.
- Optionally, click the Stretch drop-down arrow and click the type of stretch you want to use.
None
No stretch will occur. The pixel values will be mapped from the possible minimum and maximum values of the data type and depth stored in the raster to the beginning and end of the color ramp.
Custom
A customized histogram stretch can be specified by clicking the Histogram button. Add breakpoints to the graph line by clicking inside the histogram. Move them to the required location to stretch particular values.
Standard Deviation
A standard deviation stretch will be applied. You can also specify the n value for the number of standard deviations to use. This method is used to emphasize how much feature values vary from the mean value; it is best when used on normally distributed data.
Histogram Equalized
A histogram equalization contrast stretch will be applied. This method is appropriate when there are a lot of pixel values that are closely grouped together.
Minimum-Maximum
The minimum and maximum values of the calculated statistics will be used as the minimum and maximum values of the color ramp; all values in between will be interpolated linearly.
Histogram Specification
Allows you to load an XML file containing a customized histogram.
Percent Clip
A stretch is applied without using the high and low extreme percent values. You need to specify the minimum and maximum percentage values to exclude from the stretch. Valid values for the minimum and maximum are 0 to 100.
Esri
The statistics and histogram are analyzed and a modified sigmoid stretch is applied. It uses an S curve to find a mean and helps to prevent pixel values to be stretched to the extremes. This method is used to provide a good overall stretch to imagery.
Sigmoid
The Sigmoid contrast stretch is designed to highlight moderate pixel values in your imagery while maintaining sufficient contrast at the extremes. It places all of the pixel values along a sigmoidal function (an S-shaped curve). The result of this is less contrast in very bright and very dark areas, and more contrast in areas between these extremes. This is an ideal stretch for almost any image and performs very well when there are clouds and water in the image.
The Strength Level determines how much of the sigmoidal function will be used in the stretch. A low value such as 1 will only use the middle portion of the curve, which tends to produce dull and faint colors. A high value such as 6 will use the entire curve, which tends to produce bold and sharp colors.
- Optionally, check Apply Gamma Stretch and type gamma values in each text box for the red, green, and blue values.
By applying the gamma stretch, you can control the overall brightness of an image.
Valid gamma values are zero and above. If the gamma value is set too low, middle tones appear too dark; however, if the gamma value is set too high, middle tones appear too light, and the raster dataset looks bleached out. Gamma changes not only the brightness but also the ratios of red to green to blue.
Gamma stretching is only valid with some renderers: None, Standard Deviation, and Minimum-Maximum.
- Optionally, scroll down and click Statistics to change the statistics controlling how your data is displayed.
Current Display Extent
Uses the statistics only from the pixels that are currently displayed in the display view. This allows for better contrast in areas where the pixel values are similar.
Each Raster Dataset
Uses the statistics from the entire raster dataset. This ensures that the same statistics are used for each raster dataset.
Image Service
Uses the statistics from the image service layer.
Custom
Uses the statistics that you specify.
- If the raster contains a background or border around the data that you want to hide, check Display Background Value and set to No Color (transparent). Optionally, you can choose a color if you want the values to appear with a specific color.
- Optionally, click the Display NoData as button to symbolize the NoData using a color.
- If you want to pan-sharpen your raster layer, check Pan-sharpening, choose the Panchromatic Image, and modify the type and weights.
- Click OK.