DocumentCode
3120844
Title
Hausdorff distance for template detection and efficient implementation on embedded systems
Author
Petracca, Gaetano ; Leoncini, Alessio ; Sangiacomo, Fabio ; Zunino, Rodolfo
Author_Institution
Dept. of Biophys. & Electron. Eng. (DIBE), Univ. of Genoa, Genoa, Italy
fYear
2010
fDate
4-7 July 2010
Firstpage
1623
Lastpage
1626
Abstract
Template selection is often a crucial task for target-detection methods; the Hausdorff distance enhances template matching by implicitly encoding the notion of resemblance. The template-matching technique proposed fits embedded electronic systems for the real-time analysis of visual scenes. Practical implementations demonstrate the approach effectiveness by means of comparative tests on several DSP-based embedded platforms.
Keywords
embedded systems; image matching; object detection; video surveillance; DSP-based embedded platforms; Hausdorff distance; target-detection methods; template detection; template-matching technique; Digital signal processing; Image edge detection; Industrial electronics; Manganese; Pixel; Shape; Training;
fLanguage
English
Publisher
ieee
Conference_Titel
Industrial Electronics (ISIE), 2010 IEEE International Symposium on
Conference_Location
Bari
Print_ISBN
978-1-4244-6390-9
Type
conf
DOI
10.1109/ISIE.2010.5637569
Filename
5637569
Link To Document