Module: Smooth

This module smooths images using one of several filters. Fitting a polynomial is fastest but does not allow a very tight fit compared to the other methods:

• Fit Polynomial: This method treats the intensity of the image pixels as a polynomial function of the x and y position of each pixel. It fits the intensity to the polynomial, A x² + B y² + C xy + D x + E y + F. This will produce a smoothed image with a single peak or trough of intensity that tapers off elsewhere in the image. For many microscopy images (where the illumination of the lamp is brightest in the center of field of view), this method will produce an image with a bright central region and dimmer edges. But, in some cases the peak/trough of the polynomial may actually occur outside of the image itself.
• Gaussian Filter: This method convolves the image with a Gaussian whose full width at half maximum is the artifact diameter entered. Its effect is to blur and obscure features smaller than the artifact diameter and spread bright or dim features larger than the artifact diameter.
• Median Filter: This method finds the median pixel value within the artifact diameter you specify. It removes bright or dim features that are much smaller than the artifact diameter.
• Smooth Keeping Edges: This method uses a bilateral filter which limits Gaussian smoothing across an edge while applying smoothing perpendicular to an edge. The effect is to respect edges in an image while smoothing other features. Smooth Keeping Edges will filter an image with reasonable speed for artifact diameters greater than 10 and for intensity differences greater than 0.1. The algorithm will consume more memory and operate more slowly as you lower these numbers.
• Circular Average Filter: This method convolves the image with a uniform circular averaging filter whose size is the artifact diameter entered. This filter is useful for re-creating an out-of-focus blur to an image.
• Smooth to Average: Creates a flat, smooth image where every pixel of the image equals the average value of the original image.

(Used only if "Gaussian Filter", "Median Filter", "Smooth Keeping Edges" or "Circular Average Filter" is selected)
Select Yes to choose an artifact diameter based on the size of the image. The minimum size it will choose is 30 pixels, otherwise the size is 1/40 of the size of the image.

Select Yes to manually enter an artifact diameter.

(Used only if choosing the artifact diameter automatically is set to "No")
Enter the approximate diameter (in pixels) of the features to be blurred by the smoothing algorithm. This value is used to calculate the size of the spatial filter. To measure distances in an open image, use the "Measure length" tool under Tools in the display window menu bar. If you click on an image and drag, a line will appear between the two endpoints, and the distance between them shown at the right-most portion of the bottom panel. For most smoothing methods, selecting a diameter over ~50 will take substantial amounts of time to process.

(Used only if "Smooth Keeping Edges" is selected)
Enter the intensity step (which indicates an edge in an image) that you want to preserve. Edges are locations where the intensity changes precipitously, so this setting is used to adjust the rough magnitude of these changes. A lower number will preserve weaker edges. A higher number will preserve only stronger edges. Values should be between zero and one. To view pixel intensities in an open image, use the pixel intensity tool which is available in any open display window. When you move your mouse over the image, the pixel intensities will appear in the bottom bar of the display window.

(Used only if Fit Polynomial is selected)
The Fit Polynomial method is the only smoothing option that can yield an output image whose values are outside of the values of the input image. This setting controls whether to limit the image intensity to the 0 - 1 range used by CellProfiler.

Select Yes to set all output image pixels less than zero to zero and all pixels greater than one to one.

Select No to allow values less than zero and greater than one in the output image.