Model of Combined IPT and NNLVQ for Classification of Healthy and Sick Broilers In Terms of Avian Influenza

Poultry meat is an important and economical protein source in providing the animal protein requirement for human nutrition. The poultry diseases such as avian influenza that are a feature of fast-spread in farms seriously threatens both the economy and human health. Avian influenza must be detected early because it spreads rapidly. Earlier detection of poultry diseases has become more possible with the development of systems combining image processing techniques (IPTs) and artificial intelligence techniques (AITs). In this study, the neural network (NN) based model using learning vector quantization (LVQ) structure is proposed for the classification of broiler chickens as healthy and sick. In the literature, seven main visual feature parameters that indicate the health status of broilers were acquired through the IPTs. The data set includes seven visual features is used for training, testing and validating process of the NNLVQ model. The classification performance of the neural network (NN) using learning vector quantization (NNLVQ) is compared with IPT concerning its efficiency and accuracy. In the training process, the NNLVQ model classifies the broilers in terms of avian influenza with an accuracy error (AE) of 0.384%. The results point out that, the IPT based application using NNLVQ is successfully classified the broilers in terms of their health conditions.