Extracting geometric data of landmarks from fluoroscopic images plays an important role in camera calibration process of a fluoroscopic-image-based surgical navigation system. Connected components labeling is the esse...Extracting geometric data of landmarks from fluoroscopic images plays an important role in camera calibration process of a fluoroscopic-image-based surgical navigation system. Connected components labeling is the essential technique for the extraction. A new fast connected components labeling algorithm was presented. The definition of upward concave set was introduced to explain the algorithm. Feasibility and efficiency of the algorithm were verified with experiments. This algorithm performs well in labeling non-upward concave set connected components and applies to landmarks labeling well. Moreover, the proposed algorithm possesses a desirable characteristic that will facilitate the subsequent processing of fluoroscopic images.展开更多
For any multiply connected domain Ω in ?2, let S be the boundary of the convex hull in H 3 of ?2Ω which faces Ω. Suppose in addition that there exists a lower bound l > 0 of the hyperbolic lengths of closed geod...For any multiply connected domain Ω in ?2, let S be the boundary of the convex hull in H 3 of ?2Ω which faces Ω. Suppose in addition that there exists a lower bound l > 0 of the hyperbolic lengths of closed geodesics in Ω. Then there is always a K-quasiconformal mapping from S to Ω, which extends continuously to the identity on ?S = ?Ω, where K depends only on l. We also give a numerical estimate of K by using the parameter l.展开更多
基金Projectof Science and Technology Committee of Shanghai Municipality(No2528(3))
文摘Extracting geometric data of landmarks from fluoroscopic images plays an important role in camera calibration process of a fluoroscopic-image-based surgical navigation system. Connected components labeling is the essential technique for the extraction. A new fast connected components labeling algorithm was presented. The definition of upward concave set was introduced to explain the algorithm. Feasibility and efficiency of the algorithm were verified with experiments. This algorithm performs well in labeling non-upward concave set connected components and applies to landmarks labeling well. Moreover, the proposed algorithm possesses a desirable characteristic that will facilitate the subsequent processing of fluoroscopic images.
基金supported by National Natural Science Foundation of China (Grant Nos. 10671004, 10831004)the Doctoral Education Program Foundation of China (Grant No. 20060001003)
文摘For any multiply connected domain Ω in ?2, let S be the boundary of the convex hull in H 3 of ?2Ω which faces Ω. Suppose in addition that there exists a lower bound l > 0 of the hyperbolic lengths of closed geodesics in Ω. Then there is always a K-quasiconformal mapping from S to Ω, which extends continuously to the identity on ?S = ?Ω, where K depends only on l. We also give a numerical estimate of K by using the parameter l.
