Objective To realize accurate localization of moving vehicles from single monocular intens ity image. Methods The new modified Hausdorff distance(M2HD) was adopted, which used dominant points instead of edge maps a...Objective To realize accurate localization of moving vehicles from single monocular intens ity image. Methods The new modified Hausdorff distance(M2HD) was adopted, which used dominant points instead of edge maps as features for mea suring similarity between image and model projection. Modified simulated anneali ng (MSA) algorithm was used to find optimum localization parameters. Res ults M2HD reduces the computational complexity, and improves the matc hing precision. Furthermore, MSA can fast find global optimum instead of getting into partial one because of its high parallel and robust performance. C onclusion Experiments confirm that the combination of MSA and M 2H D can effectively localize the vehicles that are changed both in translation and rotation展开更多
文摘Objective To realize accurate localization of moving vehicles from single monocular intens ity image. Methods The new modified Hausdorff distance(M2HD) was adopted, which used dominant points instead of edge maps as features for mea suring similarity between image and model projection. Modified simulated anneali ng (MSA) algorithm was used to find optimum localization parameters. Res ults M2HD reduces the computational complexity, and improves the matc hing precision. Furthermore, MSA can fast find global optimum instead of getting into partial one because of its high parallel and robust performance. C onclusion Experiments confirm that the combination of MSA and M 2H D can effectively localize the vehicles that are changed both in translation and rotation
