针对仅采用局部或全局信息无法快速准确分割灰度不均匀图像的问题,提出了一种基于局部和全局信息的自适应水平集图像分割模型。首先,利用图像局部信息和全局信息建立局部能量项和全局能量项,并且利用演化曲线轮廓内外小邻域的灰度均值差作为自变量,建立了权重函数模型,实现了局部能量项和全局能量项之间权重的自适应调整,提高了模型分割灰度不均匀图像的效率和准确性。其次,提出了一种新的能量惩罚项,避免了水平集函数的重新初始化,增强了数值计算的稳定性。最后,为验证模型的优越性,将模型与CV模型、LBF模型和LGIF模型进行了对比,并通过分割时间、迭代次数以及相似度等指标对分割结果进行了客观、定量分析。最终结果表明: 该模型不但对初始轮廓具有较高鲁棒性,而且对灰度不均匀图像具有较高的分割准确性与分割效率。

In view of the problem that only using local or global intensity information cannot quickly and accurately segment images with intensity inhomogeneity, an adaptive level set model based on local and global intensity information is proposed for image segmentation. Firstly, by using local and global intensity information of image to establish the local and global energy term, and using intensity difference between the inner and the outer contour of the small neighborhood to establish weighting function, we realize adaptive adjustment of the weight between the local and global energy term, and greatly improve the efficiency and accuracy of the segmentation result. Secondly, we propose a novel energy penalty term, which avoids the re-initialization of the level set function and enhances the stability of the numerical calculation. Finally, in order to verify the superiority of the proposed model, we compare the proposed model with CV model, LBF model and LGIF model and make an objective and quantitative analysis by using the time of segmentation, the number of iterations and the similarity value. The final results show that the proposed model not only has high robustness to the initial contours, but also has high segmentation accuracy and segmentation efficiency for the images with intensity inhomogeneity.