DocumentCode :
3252061
Title :
Non-invasive lizard identification using signature curves
Author :
Li, Jim S Jimmy ; Tohl, Damian ; Randhawa, Sharmil ; Shamimi, Leili ; Bull, Michael C.
Author_Institution :
Sch. of Comput. Sci., Flinders Univ., Adelaide, SA, Australia
fYear :
2009
fDate :
23-26 Jan. 2009
Firstpage :
1
Lastpage :
5
Abstract :
A novel non-invasive method is proposed to help identify the endangered Pygmy Bluetongue Lizard. This would be preferable to the commonly used, invasive, toe-clipping method, which could be unreliable if the lizard was to lose a toe or foot naturally. Each lizard has a unique and permanent scale pattern which can be used to identify individual lizards. The proposed method involves a novel technique to derive a signature curve from the Hough transform output of the lizard scale pattern such that the signature curve is scale and shift invariant. It has been found that a lizard can be successfully identified by comparing its signature curve with those signature curves of all registered lizards stored in a database.
Keywords :
Hough transforms; pattern recognition; zoology; Hough transform output; endangered pygmy bluetongue lizard; noninvasive lizard identification; scale pattern; signature curves; toe clipping method; Adaptive equalizers; Biology; Computer science; Databases; Digital images; Foot; Head; Histograms; Image edge detection; Mathematics; Hough transform; lizard identification; non-invasive; signature curve;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
TENCON 2009 - 2009 IEEE Region 10 Conference
Conference_Location :
Singapore
Print_ISBN :
978-1-4244-4546-2
Electronic_ISBN :
978-1-4244-4547-9
Type :
conf
DOI :
10.1109/TENCON.2009.5395852
Filename :
5395852
Link To Document :
بازگشت