Classification of Poultry Birds Based on Health State Using Convolutional Neural Network

Abstract

Monitoring the health of poultry birds is essential for ensuring the productivity and safety of poultry products. The conventional methods are often time-consuming, prone to errors, and ineffective at early disease detection. The drawbacks associated with the conventional methods often results to significant financial loss and increases disease spread within flocks, thereby impacting food productivity and safety. Addressing these challenges requires innovative solutions that improve the efficiency and accuracy of poultry health management. This study is part of a research work aimed at addressing the challenges of digitalized method of poultry bird’s disease classification. To solved this, a classification of poultry birds based on health state using convolutional neural network model was developed, technique such as deep learning was used to analyzes diverse dataset of annotated images of birds with health conditions, for proper datasets classification, a convolutional neural network (CNN) was employed, the model as designed can accurately classify the health system of poultry birds from images, evaluate the performance of the developed model in terms of accuracy, precision, recall and F1 score. The model is embedded with a user-friendly interface and this was achieved through computer vision-based techniques, the interface enable users to upload images and result of different diseases as analyzed displayed