Article Preview
Top1. Introduction
Poultry farm health management is a significant aspect of the production of poultry. Many diseases in Livestock are mainly due to infection-causing agents that will extend throughout the bird's flock rapidly due to less production and subsequent death of the birds. Deficiencies of essential nutrients in feed, such as vitamins and minerals, will lead to less production. The poultry farm health management's primary keys are 1) early detection of disease; 2) if not detected, then prevention of disease; 3) if detected with the disease, advance treatment of disease and 4) Biosecurity management should be maintained regularly. The important fundamental concept is detecting disease and perceives prevention that will spread rapidly in the flock, and if the bird is identified as diseased, then treat them accordingly. The primary step of Biosecurity management should be taken care. This principle theme is of a flock health monitoring system, farm biosecurity practices, and vaccination schedule to keep a check on not good poultry flock in a farm with intelligent systems.
To recognize the disease in Livestock, an intelligent system should be designed to distinguish the diseased bird (Chedad, 2001). In this work, the basis of classification was done by using the coughing sound of the pig by discriminating other noises such as environmental sounds. The algorithm was conceptualized on Probalistic Neural Network with 91.9% accuracy. The analysis of sounds has been there for evaluation of the emotional state of an animal as suggested by many researchers. For an intelligent system to be designed, different intelligent classifiers were used by many researchers, some of which included neural networks (Boddy, 1994; Azar, 2013), decision tree (Parsons & Jones, 2000; Azar et al., 2013), Linear Discriminant Analysis (Xanthopoulos et al., 2013), Support vector machine (Azar & El-Said, 2014) and Spiking Neural network (Mohanty, 2020). The work proposed here is based on Neural Network Disease Recognition (NNDR) i.e. Adaptive Neuro-Fuzzy Inference System is used to classify healthy and unhealthy bird on the farm. These poultry birds are affected by four types of diseases 1) metabolic and nutritional diseases 2) infectious diseases 3) parasitic diseases, and 4) behavioral diseases.
The metabolic diseases and nutritional diseases can be controlled by regular feeding and watering of Livestock, which involves proper monitoring. The monitoring part is discussed in the earlier section, which is both local as well as central. The farm usually gets faster affected by the poultry's infectious diseases, which in turn give loss in production of poultry bird. With the contact of intermediate vector or direct exposure of infections, the parasitic disease gets the Livestock. The abnormal behavior of bird can cause an injury and may lead to cannibalism (aggressive pecking). So, for recognizing bird diseases, the following steps should be taken, as mentioned in Fig. 3. The disease of the four poultry bird types as mentioned in Table I is bronchitis infection in the experiment, is analyzed. The respiratory distress symptoms are sneezing and coughing, which can be easily analyzed through their vocalization. The subsequent medical help in time can be done by adding antibiotic i.e. (Tylosin) to the feed for 10 days or erythromycin in the drinking water for the same period. Not diagnosed at the right time can lead to eradication of Livestock from farm.