RUMICAM: A New Device for Cattle Rumination Analysis
Author(s) -
Gilberto de Jesus de Oliveira,
Milena Carmona,
Julia Pistori,
Patricia De Oliveira,
R. Mateus,
Geazy Vilharva Menezes,
Vanessa Weber,
Cleonice Le Bourlegat,
Hemerson Pistori
Publication year - 2020
Language(s) - English
Resource type - Conference proceedings
DOI - 10.5753/wvc.2020.13487
Subject(s) - rumination , metric (unit) , computer science , artificial intelligence , deep learning , software , support vector machine , livestock , machine learning , psychology , engineering , cognition , ecology , biology , operations management , neuroscience , programming language
Rumination may reveal important behavioral aspects of livestock animals and has been increasingly studied using new sensors technologies. In this work a new device was developed to collect close-up videos from the animal mouth during the rumination period. Using shallow and deep machine learning techniques, a software that classifies the basic mouth movements from these images has also been developed. A baseline performance for this equipment has been established using the Fscore metric. SVM achieved the highest F-score of 79.3% for the shallow learning approach. The best F-score using deep learning was 75% using VGG16.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom