作者: Max W. K. Law , Albert C. S. Chung
DOI: 10.1007/978-3-540-88693-8_27
关键词:
摘要: This paper proposes a novel curvilinear structure detector, called Optimally Oriented Flux (OOF). OOF finds an optimal axis on which image gradients are projected in order to compute the gradient flux. The computation of is localized at boundaries local spherical regions. It avoids considering closely located adjacent structures. main advantage its robustness against disturbance induced by objects. Moreover, analytical formulation introduces no additional load as compared calculation Hessian matrix widely used for detection. experimentally demonstrated that delivers accurate and stable detection responses under interference structures well noise.