作者: Dorin Comaniciu , Bogdan Georgescu , R. Bharat Rao , Alok Gupta , Xiang Sean Zhou
DOI:
关键词:
摘要: A detection framework that matches anatomical structures using appearance and shape is disclosed. training set of images are used in which object shapes or annotated the images. second represents negative examples for such structures, i.e., containing no objects structures. classification algorithm trained on sets to detect a structure at its location. The matched counterpart can provide details about structure's appearance.