DocumentCode :
2870001
Title :
Fusing 3D information for crop/weeds classification
Author :
Sáchez, A.J. ; Marchant, J.A.
Author_Institution :
Univ. Politecnica de Valencia, Spain
Volume :
4
fYear :
2000
fDate :
2000
Firstpage :
295
Abstract :
The main goal of this work is to construct a 3D-world map to help to distinguish between crop and weeds and finally to permit the location of crop and weeds to apply a variable treatment. To solve the problem of recovering a 3D-local map, we use a motion technique with the possibility of changing zoom. Every independent local map will be fused into the world map. The crop/weeds classification will eventually be performed using the height of the plants among other features like position, texture, shape, etc
Keywords :
active vision; agriculture; image classification; sensor fusion; 3D information fusion; 3D-local map; 3D-world map construction; active vision; crops classification; motion technique; weeds classification; Cameras; Crops; Feature extraction; Focusing; Gaussian processes; Image reconstruction; Remotely operated vehicles; Robot vision systems; Shape; Spraying;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition, 2000. Proceedings. 15th International Conference on
Conference_Location :
Barcelona
ISSN :
1051-4651
Print_ISBN :
0-7695-0750-6
Type :
conf
DOI :
10.1109/ICPR.2000.902917
Filename :
902917
Link To Document :
بازگشت