A modified image matching algorithm based on robust Hausdorff distance - Details

Author：

Wu, Qiang (Wu, Qiang.) | Wu, Xuefeng (Wu, Xuefeng.) | Li, Xuwen (Li, Xuwen.) | Jia, Kebin (Jia, Kebin.) (Scholars：贾克斌)

Indexed by：

EI Scopus

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.　Copyright　©　by　HIGH　TECHNOLOGY　LETTERS　PRESS.

Keyword：

Image matching Geometry Statistics

Author Community：

[ 1 ] [Wu, Qiang]College of Electronic Information and Control Engineering, Beijing University of Technology, Beijing 100124, China
[ 2 ] [Wu, Xuefeng]College of Electronic Information and Control Engineering, Beijing University of Technology, Beijing 100124, China
[ 3 ] [Li, Xuwen]College of Life Science and Bio-Engineering, Beijing University of Technology, Beijing 100124, China
[ 4 ] [Jia, Kebin]College of Electronic Information and Control Engineering, Beijing University of Technology, Beijing 100124, China