Otsu Segmentation

Segment an image using Otsu’s method for automated threshold detection. The threshold is selected to maximize the inter-region variance and minimize the intra-region variance. Returns optimal threshold values and binary image maps.

Input Parameters


Arbitrary bit-depth images with any number of channels.

Type: Image File (net.imagej.Dataset)



The channel index. Each channel of each source image is processed separately.

Type: Integer (int)


The optimal threshold value which maximizes inter-class variance:

ωa(t) ωb(t) [μa(t) - μb(t)]2

Here, t is the threshold value, ωa(t) and ωb(t) are the number of pixels in each class, and μa(t) and μb(t) are the mean pixel values of each class.

Type: Numeric (double)

Threshold Map

Binary images of each image channel segmented according to the computed optimal threshold value.

Type: Image File (net.imagej.Dataset)


Nobuyuki Otsu (1979). “A threshold selection method from gray-level histograms”. IEEE Trans. Sys., Man., Cyber. 9 (1): 62–66. doi:10.1109/TSMC.1979.4310076