Title :
3D cow identification in cattle farms
Author :
Arslan, Ahmet Cumhur ; Akar, Mehmet ; Alagoz, Fatih
Abstract :
The increased consumption of the society necessiates increased use of automation in larger animal farms which require new monitoring and tracking systems in order to observe the general health and instantaneous state of the herds. In this paper, we propose a Kinect based system, that is capable of identifying individual animals in a cattle farm from their 3D visual appearences. Since the proposed solution depends only on the shape information, animals with the same color are recognized successfully even in poor lighting conditions. Furthermore, as no markers or external devices such as tags are utilized, the natural behavior of the animals is affected in a minimal way.
Keywords :
automation; identification; image recognition; image sensors; monitoring; tracking; 3D cow identification; 3D visual appearences; Kinect based system; cattle farms; color recognition; external devices; lighting conditions; markers devices; monitoring systems; natural behavior; shape information; tracking systems; Conferences; Cows; Global Positioning System; Monitoring; Signal processing; Three-dimensional displays; 3D; Kinect; cattle; farm; identification;
Conference_Titel :
Signal Processing and Communications Applications Conference (SIU), 2014 22nd
Conference_Location :
Trabzon
DOI :
10.1109/SIU.2014.6830487