Deep Learning-Based Cattle Vocal Classification Model and Real-Time Livestock Monitoring System with Noise Filtering

Abstract

Simple Summary In the application of artificial intelligence and advanced sound technologies in animal sound classification, certain challenges are still faced, such as the disruptions of background noise. To address this problem, we propose a web-based and real-time cattle monitoring system for evaluating cattle conditions. The system contained a convolutional neural network (CNN) for classifying cattle vocals and removing background noise as well as another CNN for behavior classification from existing datasets. The developed model was applied to cattle sound data obtained from an on-site monitoring system through sensors and achieved a final accuracy of 81.96% after the sound filtering. Finally, the model was deployed on a web platform to assist farm owners in monitoring the conditions of their livestock. We believe that our study makes a significant contribution to the literature because it is the first attempt to combine CNN and Mel-frequency cepstral coefficients (MFCCs) for real-time cattle sound detection and a corresponding behavior matching. The priority placed on animal welfare in the meat industry is increasing the importance of understanding livestock behavior. In this study, we developed a web-based monitoring and recording system based on artificial intelligence analysis for the classification of cattle sounds. The deep learning classification model of the system is a convolutional neural network (CNN) model that takes voice information converted to Mel-frequency cepstral coefficients (MFCCs) as input. The CNN model first achieved an accuracy of 91.38% in recognizing cattle sounds. Further, short-time Fourier transform-based noise filtering was applied to remove background noise, improving the classification model accuracy to 94.18%. Categorized cattle voices were then classified into four classes, and a total of 897 classification records were acquired for the classification model development. A final accuracy of 81.96% was obtained for the model. Our proposed web-based platform that provides information obtained from a total of 12 sound sensors provides cattle vocalization monitoring in real time, enabling farm owners to determine the status of their cattle.