为了准确分离目标区域,提出了一种基于贪婪Snake与多尺度图像增强相结合的新的轮廓分离模型.首先对输入的图像给定一个初始轮廓,然后应用多尺度分析进行图像增强,之后在不同的尺度中运用贪婪Snake模型进行轮廓的分离,最终分离出理想的目标轮廓.实验证明该方法提高了分割的精度.
Image segmentation is an important issue in the field of imaging and image processing.In order to separate target area accurately,a new method was proposed based on greedy Snake and muti-scale analysis to separate contour.An initial contour was given,image was enhanced by muti-scale analysis.The contour was then separated using the greedy snake model on different scales,to obtain an ideal contour.Experiments showed that this method improved segmentation accuracy.