A Classification Scheme Based on Directed Acyclic Graphs for Acoustic Farm Monitoring

Abstract

Intelligent farming as part of the green revolution is advancing the world of agriculture in such a way that farms become evolving, with the scope being the optimization of animal production in an eco-friendly way. In this direction, we propose exploiting the acoustic modality for farm monitoring. Such information could be used in a stand-alone or complimentary mode to monitor constantly animal population and behavior. To this end, we designed a scheme classifying the vocalizations produced by farm animals. More precisely, we propose a directed acyclic graph, where each node carries out a binary classification task using hidden Markov models. The topological ordering follows a criterion derived from the Kullback-Leibler divergence. During the experimental phase, we employed a publicly available dataset including vocalizations of seven animals typically encountered in farms, where we report promising recognition rates outperforming state of the art classifiers.