Straehle C N, Köthe U, Knott G, Hamprecht F A
University of Heidelberg, Heidelberg, Germany HCI, Speyerer Strasse 6, D-69115 Heidelberg.
Med Image Comput Comput Assist Interv. 2011;14(Pt 1):653-60. doi: 10.1007/978-3-642-23623-5_82.
Interactive segmentation algorithms should respond within seconds and require minimal user guidance. This is a challenge on 3D neural electron microscopy images. We propose a supervoxel-based energy function with a novel background prior that achieves these goals. This is verified by extensive experiments with a robot mimicking human interactions. A graphical user interface offering access to an open source implementation of these algorithms is made available.
交互式分割算法应在数秒内做出响应，且只需最少的用户指导。这对于三维神经电子显微镜图像来说是一项挑战。我们提出了一种基于超体素的能量函数，并带有一种新颖的背景先验知识，以此实现这些目标。通过一个模仿人类交互的机器人进行的大量实验验证了这一点。我们提供了一个图形用户界面，可用于访问这些算法的开源实现。
