From the point of view of information divergence, this paper proposed a new weighting method of 3D-Otsu, which could obtain a precise segmentation result for inhomogeneous image. Since the three dimensions in 3D-Otsu method played different roles in segmentation of images, could obtain better results of segmentation by changing the weighting factors of the three dimensions in 3D-Otsu method. According to the characteristic of weighting method, the 3D-Otsu could be decomposed into three one-dimensions. So, this method could speed up the calculation and lower the computational complexity and made the weighting method more practical. With this method, this paper obtained an optimized segmentation result.%从信息散度角度出发,提出一种新的三维Otsu加权方法,对分布不均匀的图像做细致分割.由于三维Otsu中三个维度在分割中所起的作用不同,可以通过调整不同维度的权因子使不同的图像都能得到很好的分割效果.最后利用加权的特点,将三维Otsu分解为三个一维Otsu,在加快运算速度、降低时间复杂度的同时得到了比较好的分割效果,使得加权方法更具有实际效用.
展开▼