A modified image matching algorithm based on robust Hausdorff distance

【摘要】 Hausdorff distance measure is one of the widely adopted feature-based image matching algorithms due to its simplicity and accuracy.However,it is considered that its robustness still needs to be improved.In this paper,various forms of original and improved Hausdorff distance(HD) and their limitations are studied.Focusing on robust Hausdorff distance(RHD),an improved RHD with an adaptive outlier point threshold selection method is proposed.Furthermore,another new form of the Hausdorff distance which possesses the merits of RHD and M-HD is prsented.Finally,a recursive algorithm is introduced to accelerate the image matching speed of Hausdorff algorithms.Extensive simulation and experiment results are presented to validate the feasibility of the proposed Hausdorff distance algorithm.更多还原

【Abstract】 Hausdorff distance measure is one of the widely adopted feature-based image matching algorithms due to its simplicity and accuracy.However,it is considered that its robustness still needs to be improved.In this paper,various forms of original and improved Hausdorff distance(HD) and their limitations are studied.Focusing on robust Hausdorff distance(RHD),an improved RHD with an adaptive outlier point threshold selection method is proposed.Furthermore,another new form of the Hausdorff distance which possesses the merits of RHD and M-HD is prsented.Finally,a recursive algorithm is introduced to accelerate the image matching speed of Hausdorff algorithms.Extensive simulation and experiment results are presented to validate the feasibility of the proposed Hausdorff distance algorithm.更多还原