Classifying broiler chicken condition using audio data

Abstract

This paper is an effort to help prevent broiler chicken mortality caused by stressful conditions. We assume a relation between broiler chicken vocalizations and stress; therefore, microphones were used to monitor a flock of birds over the course of their lifetime (approximately 65 days). A noise removal method based on spectral oversubtraction was developed to filter out the significant fan and heater noise and shown to be very effective. Then, a radar processing technique was employed to count the number of vocalizations. It was found that the number of vocalizations is an effective technique for detecting stressful conditions, easily classifying the gathered test cases using a threshold classifier with perfect accuracy. Therefore, we conclude that this system could easily be adapted into an effective, inexpensive poultry flock monitoring tool, and the methods developed here could be applied to other similar monitoring applications.