Local Distribution  
 Top  Previous  Next

 
Menu:   Processing > Threshold ->Automatic->Minimize error  
Script   threshDistrLocal  

Use this command to automatically separate objects of interest from the background by using the adaptive error minimization technique. The method analyzes the histogram of the local neighborhood surrounding each pixel and tries to match it to a histogram in which two bell curves (one of light objects and the other of dark objects) overlap. The threshold is locally set to that which minimizes the classification error at the overlap.

If you choose this command, the following options will become available:

Input
 
Displays the input image frame number. If you want to apply thresholding to another image, type or select the corresponding value.  

Output
 
Displays the number of the frame in which the segmented binary image will be created. Depending on the Preferences. ImageWarp will set it either to the first available value or to the Input frame number. Type or select another value if you want the output image to be created in a different frame.  

Channel
 
Lets you select the color channel that will be used for the histogram analysis.  
 
Click Luminance to separate objects of interests based on the intensity distribution.  
Click Hue to separate objects of interests based on the color pigment distribution.  
Click Saturation to separate objects of interests based on the color saturation distribution.  
 
This option is disregarded for grayscale images.  
 
Window
 
Lets you select the neighborhood size for the histogram collection. For better results this value should be comparable with the size of the largest object of interest in the image.  

Invert
 
Select this check box to invert the operation. The pixels whose values are equal or above the calculated threshold level will be treated as background ones, while the pixels with values below the will be treated as foreground ones.  
 
Ignore B&W
 
Select this check box to exclude saturated black and white pixels (minimum and maximum pixel values for the given type of image) from the calculation.