马尔科夫随机场在生物医学图像分割中的应用

Application of Markov random fields in segmentation of image

目的：建立一种图像分割方法，以适应精确的生物医学图像分析。方法：根据马尔科夫随机场图像模型，利用最大后验概率准则（ Ｍ Ａ Ｐ），提出一种迭代松弛分割算法。结果：在用于生物医学图像以及 Ｃ Ｔ重建图像的分析测量中得到了较为理想的实验结果。结论：该算法与经典的图像分割算法相比具有分割精度高、可以进行多区域图像分割的特点。

Objective:To seek a method of segmentation for biomedical image. Methods:Based on Markov random fields model of noise, a iteration algorithm was presented by using maximum a posteriori (MAP) criterion. Results:Some ideal experimental results showed that the algorithm had a vast range of application prospect in analysis and measurement of medical cell image and reconstruction of CT image. Conclusion: Compared with other classical algorithms, this algorithm has the characteristics of high precision for segmentation and multi region segmentation.