| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 6541030 | Computers and Electronics in Agriculture | 2013 | 10 Pages |
Abstract
⺠A supervised-learning approach to sow-activity classification is introduced. ⺠Acceleration patterns are used, time dependencies between measurements are ignored. ⺠The method is able to classify measurements recorded at an instance of time. ⺠The method allows the online monitoring of animals in (near) real-time. ⺠Experimental results outperform previous work by a considerable margin.
Keywords
Related Topics
Physical Sciences and Engineering
Computer Science
Computer Science Applications
Authors
Hugo Jair Escalante, Sara V. Rodriguez, Jorge Cordero, Anders Ringgaard Kristensen, Cécile Cornou,