基于最小能量的图像分割算法

Image Segmentation Algorithm Based on Minimum Energy Principle

文中对图像分割进行了研究,将图像分割过程看成一获取信息的过程,通过最大化此信息量,导出一分割图像的能量泛函,最小化能量泛函即可得到分割结果。在最小化能量函数时,运用最陡梯度下降法导出一曲线进化方程,并用水平集方法来表示此曲线,利用快速水平集方法来进行数值求解。最后的仿真结果表明该算法很有效。

A novel algorithm for image segmentation is presented. We take into account the image segmentation as a process of obtaining information and get an energy function via maximum entropy principle. We can get the segmentation result through minimizing the energy function. The energy function is minimized by using gradient-descent methods. which leads to a curve evolution equation that segments the image. We use level set method to represent the curve evolution equation and solve the equation utilizing fast level set method. The experimental results indicate the method is fine.