Based on the coded and non-coded targets, the targets are extracted from the images according to their size, shape and intensity etc., and thus an improved method to identify the unique identity（D） of every coded ta...Based on the coded and non-coded targets, the targets are extracted from the images according to their size, shape and intensity etc., and thus an improved method to identify the unique identity（D） of every coded target is put forward and the non-coded and coded targets are classified. Moreover, the gray scale centroid algorithm is applied to obtain the subpixel location of both uncoded and coded targets. The initial matching of the uncoded target correspondences between an image pair is established according to similarity and compatibility, which are based on the ID correspondences of the coded targets. The outliers in the initial matching of the uncoded target are eliminated according to three rules to finally obtain the uncoded target correspondences. Practical examples show that the algorithm is rapid, robust and is of high precision and matching ratio.展开更多
Deep-learning methods provide a promising approach for measuring in-vivo knee joint motion from fast registration of two-dimensional(2D)to three-dimensional(3D)data with a broad range of capture.However,if there are i...Deep-learning methods provide a promising approach for measuring in-vivo knee joint motion from fast registration of two-dimensional(2D)to three-dimensional(3D)data with a broad range of capture.However,if there are insufficient data for training,the data-driven approach will fail.We propose a feature-based transfer-learning method to extract features from fluoroscopic images.With three subjects and fewer than 100 pairs of real fluoroscopic images,we achieved a mean registration success rate of up to 40%.The proposed method provides a promising solution,using a learning-based registration method when only a limited number of real fluoroscopic images is available.展开更多
基金The National Natural Science Foundation of China(No50475041)
文摘Based on the coded and non-coded targets, the targets are extracted from the images according to their size, shape and intensity etc., and thus an improved method to identify the unique identity（D） of every coded target is put forward and the non-coded and coded targets are classified. Moreover, the gray scale centroid algorithm is applied to obtain the subpixel location of both uncoded and coded targets. The initial matching of the uncoded target correspondences between an image pair is established according to similarity and compatibility, which are based on the ID correspondences of the coded targets. The outliers in the initial matching of the uncoded target are eliminated according to three rules to finally obtain the uncoded target correspondences. Practical examples show that the algorithm is rapid, robust and is of high precision and matching ratio.
基金sponsored by the National Natural Science Foundation of China(31771017,31972924,81873997)the Science and Technology Commission of Shanghai Municipality(16441908700)+3 种基金the Innovation Research Plan supported by Shanghai Municipal Education Commission(ZXWF082101)the National Key R&D Program of China(2017YFC0110700,2018YFF0300504,2019YFC0120600)the Natural Science Foundation of Shanghai(18ZR1428600)the Interdisciplinary Program of Shanghai Jiao Tong University(ZH2018QNA06,YG2017MS09).
文摘Deep-learning methods provide a promising approach for measuring in-vivo knee joint motion from fast registration of two-dimensional(2D)to three-dimensional(3D)data with a broad range of capture.However,if there are insufficient data for training,the data-driven approach will fail.We propose a feature-based transfer-learning method to extract features from fluoroscopic images.With three subjects and fewer than 100 pairs of real fluoroscopic images,we achieved a mean registration success rate of up to 40%.The proposed method provides a promising solution,using a learning-based registration method when only a limited number of real fluoroscopic images is available.
