Otsu's Method Algorithm

Otsu's method is an algorithm for automatically choosing the best threshold to convert a grayscale image into a black-and-white image, separating its pixels into a foreground class and a background class. It works directly from the image's brightness histogram, testing every possible threshold value and, for each one, calculating how much the two resulting classes' brightness values vary internally; the algorithm selects the threshold that minimizes this within-class variance, which is mathematically equivalent to maximizing the variance between the two classes, so the chosen cut point separates the two groups as cleanly as the data allows. Nobuyuki Otsu published the method in 1979 in IEEE Transactions on Systems, Man, and Cybernetics, and it remains the standard automatic thresholding technique used across image processing and computer vision whenever a simple, data-driven binary split of an image is needed.