Wolfram Language & System 10.4 (2016)|Legacy Documentation

This is documentation for an earlier version of the Wolfram Language.View current documentation (Version 11.2)


creates a binary image from image by replacing values above the mean of the range-r neighborhood with 1 and others with 0.

replaces values above with 1 and others with 0, where and are the local mean and standard deviation.

Details and OptionsDetails and Options

  • Local adaptive binarization determines the binarization thresholds locally and is typically used for segmenting the foreground of an image with nonuniform illumination or background.
  • LocalAdaptiveBinarize[image,r] is equivalent to LocalAdaptiveBinarize[image,r,{1,0,0}].
  • Using radius r, local mean and standard deviation are computed over blocks centered on each pixel.
  • LocalAdaptiveBinarize works with 2D and 3D images, starting the binarization by converting multichannel and color images into grayscale.
  • The following options can be used:
  • Padding"Fixed"what padding scheme to use
    PerformanceGoal$PerformanceGoalwhat aspect of performance to optimize
Introduced in 2014
| Updated in 2016