本文算法对原图像进行快速Mallat小波分解得到骨干图后,利用其高频子带梯度信息优化边的权重,并在概率阈值的准则下对争议区域做进一步划分,最后把最大到达概率所在类的标签赋予未标定顶点,并扩展到原图像,得到分割边界.用微软GrabCut分割数据库图像和实际DR图像对该算法进行了验证,该算法能快速而有效地分割出特定的图像,适用于DR图像的分割,为进行组织增强和进一步提高DR图像质量打下基础.

1. Pohle R,Toennies KD.Segmentation of medical images using adaptive region growing[C].Proc SPIE Medical Imaging,2001,2(27):1337-1346.
2. Grau V,Mewes A,Alcaniz M,et al.Improved watershed transform for medical image segmentation using prior information[J].IEEE Transactions on Medical Imaging,2004,23 (4):447-458,145-148.
3. 朱付平,田捷,林瑶,等.基于Level Set方法的医学图像分割[J].软件学报,2002,13(9):1866-1872.
4. Leo Grady.Random walks for image segmentation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2006,28(11):1768-1783.
5. Grady L,Funka-Lea G.Multi-Label Image Segmentation for Medical Applications Based on Graph-Theoretic Electrical Potentials[C].Proc Workshop Computer Vision and Math.Methods in Medical and Biomedical Image Analysis,2004,3117:230-245.
6. Duchenne O,Audibert J,Keriven R,et al.Segmentation by transduction[C].Computer Vision and Pattern Recognition,2008:1-8.
7. Zhu Xiaojin.Lafferty John.Harmonic mixtures:combining mixture models and graph-based methods for inductive and scalable semi-supervised learning[C].Proceedings of the 22nd International Conference on Machine Learning,2005:1057-1064.
8. Mallat SG.A theory for multiresolution signal decomposition:The wavelet representation[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1989,11(7):674-693.
9. Juan Liu,Moulin P.Information-Theoretic Analysis of Interscale and Intrascale Dependencies Between Image Wavelet Coefficients[J].IEEE Transactions on Image Processing,2001,10(11):1647-1658.