摘要
提出了基于像素邻域平均灰度的改进的一维Otsu多门限分割方法 ,兼有准确性和快速性的优点 ,并应用于B超图像 ,实现了多个目标的自动分割 ,与普通的一维Otsu法相比 ,具有更好的分割效果。
In this paper, A one-dimentional Otsu multithresholding algorithm based on the average grey value of the pixel′s neighborhood is proposed, which has the merit of both speed and accuracy. It is applied to the multi-object automatic segmentation of B-scan ultrasonic images and gets a better result than one-dimentional Otsu method.
出处
《电子工程师》
2003年第11期49-52,共4页
Electronic Engineer
