Chapter 14 - Livestock health monitoring using a smart IoT-enabled neural network recognition system

Abstract

A poultry farm monitoring system along with a livestock monitoring system is a principal system for investigating the status of bird health by collecting biological traits such as their uttered sounds. This theme combines the Internet of Things (IoT) with nonintrusive and reliable wearable sensing technology. In recent developments, machine learning (ML) or artificial neural networks (ANNs) have been well applied and recognized as an effective tool for a range of complicated scenario analyses in real time, including healthcare sector applications. This will help to alleviate problems typically suffered and faced by medical researchers in these fields by saving time for practitioners by providing unbiased results. In this chapter we discuss the utilization of analytics learning and neural network usage toward clinical concerns in health care using the IoT. ANN is a prime research domain with recent deployment of computational sophistication in hardware and software in several application domains with highly complicated computing scenarios. The healthcare sector is one area which is capable of automation to save time and that is subjective by nature. Therefore, ML- and ANN-based simulations generate unbiased outcomes. This chapter describes an IoT-structured wearable sensing platform with the inclusion of an audio feature and temperature of the livestock. In particular, the secure audio-wellbeing features are incorporated into the platform to spontaneously examine and conclude using voice information from the livestock for recognition of diseased birds. One month of long-term recognition experimentation analysis was performed where the recognition accuracy of the onset of disease bird was about 91% using a spiking neural network (SNN).The recognition accuracy of SNN in this regard is better than the performance of an ANN. A sequence of steps was taken in connection with a specific event that occurred and involved in examining the interrelationship across the central monitoring unit and the local monitoring unit using the IoT by utilizing the bird voice features, bird temperature, and room temperature and humidity.