Improved approach for object location under affine transformation using the Hausdorff distance

An improved approach to locating an object that has undergone an affine transformation is presented. This approach is based on the so-called "smallest square box distance transform" introduced in this paper. Experimental results show that the proposed approach distinctly outperforms Rucklidge's fast methods.

[1]  William Rucklidge,et al.  Efficiently Locating Objects Using the Hausdorff Distance , 1997, International Journal of Computer Vision.

[2]  Dong-Gyu Sim,et al.  Nonparametric hierarchical Hausdorff distance matching algorithm , 2000 .

[3]  Daniel P. Huttenlocher,et al.  A multi-resolution technique for comparing images using the Hausdorff distance , 1993, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[4]  Daniel P. Huttenlocher,et al.  Comparing Images Using the Hausdorff Distance , 1993, IEEE Trans. Pattern Anal. Mach. Intell..