Article (Scientific journals)
HOW MANY HIPPOS (HOMHIP): Algorithm for automatic counts of animals with infra-red thermal imagery from UAV
Lhoest, Simon; Linchant, Julie; Quevauvillers, Samuel et al.
2015In International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XL-3/W3
Peer Reviewed verified by ORBi
 

Files


Full Text
isprsarchives-Lhoest-2015.pdf
Publisher postprint (1.2 MB)
Download

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Hippos; UAV; Infra-red; Thermal-imagery; Automatic count; Algorithm; QGIS
Abstract :
[en] The common hippopotamus (Hippopotamus amphibius L.) is part of the animal species endangered because of multiple human pressures. Monitoring of species for conservation is then essential, and the development of census protocols has to be chased. UAV technology is considering as one of the new perspectives for wildlife survey. Indeed, this technique has many advantages but its main drawback is the generation of a huge amount of data to handle. This study aims at developing an algorithm for automatic count of hippos, by exploiting thermal infrared aerial images acquired from UAV. This attempt is the first known for automatic detection of this species. Images taken at several flight heights can be used as inputs of the algorithm, ranging from 38 to 155 meters above ground level. A Graphical User Interface has been created in order to facilitate the use of the application. Three categories of animals have been defined following their position in water. The mean error of automatic counts compared with manual delineations is +2.3% and shows that the estimation is unbiased. Those results show great perspectives for the use of the algorithm in populations monitoring after some technical improvements and the elaboration of statistically robust inventories protocols.
Disciplines :
Environmental sciences & ecology
Author, co-author :
Lhoest, Simon  ;  Université de Liège > Ingénierie des biosystèmes (Biose) > Laboratoire de Foresterie des régions trop. et subtropicales
Linchant, Julie ;  Université de Liège > Ingénierie des biosystèmes (Biose) > Gestion des ressources forestières et des milieux naturels
Quevauvillers, Samuel ;  Université de Liège > Ingénierie des biosystèmes (Biose) > Gestion des ressources forestières et des milieux naturels
Vermeulen, Cédric ;  Université de Liège > Ingénierie des biosystèmes (Biose) > Laboratoire de Foresterie des régions trop. et subtropicales
Lejeune, Philippe ;  Université de Liège > Ingénierie des biosystèmes (Biose) > Gestion des ressources forestières et des milieux naturels
Language :
English
Title :
HOW MANY HIPPOS (HOMHIP): Algorithm for automatic counts of animals with infra-red thermal imagery from UAV
Publication date :
2015
Journal title :
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
ISSN :
1682-1750
eISSN :
2194-9034
Publisher :
Copernicus
Special issue title :
ISPRS Geospatial Week 2015
Volume :
XL-3/W3
Peer reviewed :
Peer Reviewed verified by ORBi
Available on ORBi :
since 19 October 2015

Statistics


Number of views
980 (82 by ULiège)
Number of downloads
1027 (18 by ULiège)

Scopus citations®
 
54
Scopus citations®
without self-citations
54
OpenCitations
 
36
OpenAlex citations
 
60

Bibliography


Similar publications



Contact ORBi