Two-Dimensional Improved Otsu Using Histogram Division-Calibration

Two-dimensional Otsu algorithm is applied widely in image processing and computer vision. However, there are many problems in the original two-dimensional Otsu algorithm. This paper proposes a novel two-dimensional Otsu improved method based on two-dimensional histogram division and calibration method on account of the problem of the improper region division in the traditional algorithm. In this paper, the two dimensional histogram is divided by histogram division and calibration method to demarcate the calculation region. On this basis, the two-dimensional Otsu method is modified to make it more reasonable. The experiments show that this improved two-dimensional Otsu method is more robust and less time-consuming than the algorithms before.