Feathered Prognostics: A Deep Dive Into Poultry Disease Classification Models

This study has introduced a pioneering endeavor in the realm of poultry farming, which aims to mitigate the burgeoning challenges of disease detection. With poultry farming on the rise, there is a concurrent escalation of the transmission of infectious maladies among poultry, posing not only economic ramifications but also a potential threat to human health. Manual inspection for poultry diseases is both labor-intensive and time-consuming. To address all these pressing concerns, the proposed work endeavors to construct a sophisticated deep-learning model that is proficient in disease detection within poultry populations, leveraging cutting-edge technologies such as CNNs in conjunction with an efficient net B3 optimizer. The developed model showcased exceptional performance metrics with a training loss of 0.524311 and an accuracy of 99.67%. Furthermore, rigorous validation and testing yielded robust results, demonstrating a validation accuracy of 96.90% and a testing accuracy of 97.52%. Through the amalgamation of state-of-the-art deep learning methodologies, this research project aspires to revolutionize disease identification in poultry.