Poultry disease detection method using deep learning technology for different breads

The project focuses on the early detection of poultry diseases using deep learning techniques, specifically Convolutional Neural Networks (CNNs), implemented in MATLAB. The system is divided into three main phases, each focusing on a different aspect of poultry behavior and health. In the first phase, Vocal Classification, the system classifies poultry vocalizations into categories such as Healthy, Newcastle disease, Bronchitis Virus, and Influenza based on predefined image datasets. The second phase, Dropping Classification, analyzes poultry droppings and classifies them into categories including Healthy, Salmonella, Coccidiosis, and Newcastle disease, again utilizing deep learning algorithms. The third phase, Behavioural Characteristics Classification, focuses on classifying the behavioural patterns of chickens into various states such as Running, Standing, Walking, Eating, Resting, and Preening. The classification outputs from each phase are provided in text format, offering easy-to-interpret results for disease diagnosis. This system leverages MATLAB's capabilities for image processing and deep learning, providing a user-friendly interface where users can select the classification type based on their requirements. By automating the identification of poultry diseases through image-based analysis, this project aims to support farmers and veterinarians in early diagnosis and prevention, enhancing poultry health management.

Keywords: Dataset, Image Processing Techniques, Deep Learning, Convolution Neural Network classification and Accuracy.