Revolutionizing Cattle Health: A Machine Learning Approach to Efficiently Predict Lumpy Disease in Cows

Diseases in livestock can have a significant impact on the economic growth of a country. and the emergence of new diseases poses a major concern for their early detection. Lumpy disease, in particular, has become increasingly common and leads to substantial economic losses for farmers. Detecting and predicting the disease at an early stage can help mitigate its spread and implement preventive measures. In line with this objective, this study focuses on the image-based classification of lumpy disease in cattle. The proposed approach involves preprocessing the images, training a Convolutional Neural Network (CNN) model on a large dataset of cattle images, and evaluating the model’s performance on a separate testing dataset. The methodology includes the utilization and comparison of three well-known CNN architectures: VGG16, VGG19, and Inception V3. The accuracy achieved by these models was 87%, 86%, and 85% respectively. Additionally, the study investigates the performance of four popular activation functions, namely ReLU, Sigmoid, Tanh, and Linear, in the fully-connected layers of the CNN model.