学术报告
Minimal Paths for Image Segmentation and Tubular Structure Tracking - 陈达 副教授（山东省人工智能研究院，国家超算济南中心）
题目：Minimal Paths for Image Segmentation and Tubular Structure Tracking
报告人：陈达 副教授
（山东省人工智能研究院，国家超算济南中心）
Abstract：
The minimal path technique was introduced about 20 years ago as an inter- active tool for object segmentation through the Eikonal PDE approach. As an important merit, the minimal paths method can find the global minimum of the geodesic active contour energy, for which the evolving curves might be trapped into unexpected local minima. Minimal paths have been used for long as an interactive tool to segment object boundary and to track tubular structure cen- terlines, both of which can be modeled as minimizing curves. The user usually provides start and end points on the image and gets the minimal path as out- put. In essence, these minimal paths are minimal geodesic curves according to some adapted metrics. They are a way to find a curve globally minimizing the geodesic active contours energy, done by solving the corresponding Eikonal PDE using the fast and efficient Fast Marching method.In contrast to find the global minimum of a simplified active contour energy, we have recently extended these methods to cover all kinds of active con- tour energy terms. Through finding geodesic paths for new metrics, we now able to estimate minimal geodesic paths according to various active contours terms, involving curvature penalization and region-based homogeneity term. We will present the mathematical background as well as concrete applications to biomedical and natural images.
时间：2019年11月29日（周五）下午14:00-15:00
地点：首都师范大学本部教二613教室
主办单位：
首都师范大学77779193永利官网
检测成像北京市高校工程研究中心
北京成像理论与技术高精尖创新中心
欢迎全体师生积极参加！
