【摘要】 针对经典的水平集算法(比如Chan-Vese模型算法)在迭代过程中要重新初始化和容易受噪声和模棱两可的边界的影响的缺点,增加一项内部能量泛函达到不需重新初始化的目的,并结合贝叶斯决策理论,利用图像先验知识,提出了一个改进的能量函数,根据符号距离函数来不断调整水平集函数的偏差。该函数是利用期望最大化算法来得到的。实验结果表明,该算法分割精度和运行准确率上都优于经典算法。更多还原

【Abstract】 Aimed at the shortcomings of the classical level set methods such as the Chan-Vese model algorithm in the iteration process to re-initialize and easily affected by noise and ambiguous boundaries,an internal energy functional is added to achieve the purpose of without re-initialization,and prior knowledge of image is combined with Bayesian decision theory to propose an improved energy function to tackle this problem by continuously rectifying the deviation of the level set function according to the signed distance function.This is achieved using an expectation-maximisation algorithm.Experimental results shows the proposed algothim is better than the classical image segmentation on precision and accuracy.更多还原