Wang Li, Li Chunming, Sun Quansen, Xia Deshen, Kao Chiu-Yen
School of Computer Science & Technology, Nanjing University of Science and Technology, China.
Med Image Comput Comput Assist Interv. 2008;11(Pt 1):384-92. doi: 10.1007/978-3-540-85988-8_46.
In this paper, we present an improved region-based active contour/surface model for 2D/3D brain MR image segmentation. Our model combines the advantages of both local and global intensity information, which enable the model to cope with intensity inhomogeneity. We define an energy functional with a local intensity fitting term and an auxiliary global intensity fitting term. In the associated curve evolution, the motion of the contour is driven by a local intensity fitting force and a global intensity fitting force, induced by the local and global terms in the proposed energy functional, respectively. The influence of these two forces on the curve evolution is complementary. When the contour is close to object boundaries, the local intensity fitting force became dominant, which attracts the contour toward object boundaries and finally stops the contour there. The global intensity fitting force is dominant when the contour is far away from object boundaries, and it allows more flexible initialization of contours by using global image information. The proposed model has been applied to both 2D and 3D brain MR image segmentation with promising results.
在本文中，我们提出了一种改进的基于区域的活动轮廓/曲面模型，用于二维/三维脑部磁共振图像分割。我们的模型结合了局部和全局强度信息的优点，这使得该模型能够应对强度不均匀性。我们定义了一个能量泛函，它包含一个局部强度拟合项和一个辅助全局强度拟合项。在相关的曲线演化中，轮廓的运动分别由所提出的能量泛函中的局部项和全局项诱导的局部强度拟合力和全局强度拟合力驱动。这两种力对曲线演化的影响是互补的。当轮廓靠近物体边界时，局部强度拟合力占主导地位，它将轮廓吸引到物体边界并最终在那里停止轮廓。当轮廓远离物体边界时，全局强度拟合力占主导地位，并且它允许通过使用全局图像信息对轮廓进行更灵活的初始化。所提出的模型已应用于二维和三维脑部磁共振图像分割，取得了有前景的结果。
