Region Statistics
The Region Statistics dialog configures the computation of a region statistic for a parent command (for example, Combine Images and Normalize Image). In this dialog you describe the statistical estimator to be computed and define the bounds of the rectangular region where it is computed. The estimator is a single value that represents the pixels inside the specified region.
This dialog is typically opened from the parent dialog using the button.
As shown below, this dialog has a different appearance depending on whether the chosen statistic uses parameters. For example, the Median statistic uses no parameters and looks like the dialog on the left, below. Conversely, the Mean - Sigma Clipped staistic has two parameters, High sigma and Low sigma, and looks like the other dialog.
This Region Statistics dialog provides both Absolute and Relative region descriptions. The absolute region uses the coordinates of a displayed image . The relative region uses percent coordinates for the size and location of the reference rectangle, which makes it independent of the image size.. In comparison the Combine Files command uses a form of this dialog that provides only the Relative option because it does not use displayed images whose size is known before it is opened.
Mira provides numerous ways to specify the region Mira uses for calculating the estimator (see Image Region Statistics for a description of the options). The dialog above shows the Mean - Sigma Clipped statistic as the selection. Compared with a simple statistics like the Mean or Median, sigma clipping requires you to specify two parameters, the High Sigma and the Low Sigma. Some other methods require 1 or 0 parameters. As a result, there may appear up to two additional text fields on the right side of the Statistic selection box. The picture below shows the selection box dropped after the Mean - Sigma Clipped statistic was selected.
The Reference Region is a rectangular area that encloses the pixels used to compute the normalization statistic. Choosing the correct region is as important as choosing the correct statistics to measure.
Use an Absolute region to specify exact edge coordinates for the region. There are two ways to choose the column and row limits of the region:
Use Cursor ROI checked: In this case, Mira ignore the column and row settings in the dialog and uses the bounds of the Image Cursor when the parent command needs to compute the normalization. This allows you to visually adjust the cursor to enclose the region you want to use.
Use Cursor ROI not checked: In this case, Mira ignore the Image Cursor and uses the specific column and row limits entered into this dialog.
You can see that the term "absolute" means that the column and row boundaries are specified by you and must be within the bounds of the target image.
Use a Relative region to specify the region as a percentage of the image dimensions. The actual column and row coordinates of the region are calculated for each image and depend on the actual image size.
The Size % preference is based upon percentage of the image size. For example, a value of 10 indicates that the normalization rectangle extends 10% of the image width or height, whichever is the case. It does not matter how large the image may be. In this example, the normalization rectangle always covers 10% of the image dimension.
The Center % preference defines the central position of the references region as a percentage of the image size. For example, a value of 50 indicates that the normalization rectangle is centered at 50% of the image width or height, whichever is the case. It does not matter how large the image may be. In this example, the normalization rectangle is always centered at 50% of the image dimension.
Image Region Statistics, Combine Images, Normalize Image, Combine Files