Over the past few years, citrus fruit production has been facing major hurdles due to various types of diseases threatening its yield and quality. Rapid disease status of these diseases is one of the prerequisites for effective management and mitigation strategies. A Convolutional Neural Network (CNN) is trained to automatically detect the diseases in citrus fruit based on image detection by designing the CNN architecture. The architecture of the Convolutional Neural Network is formed by training it with a vast dataset involving labeled both healthy and diseased citrus fruits. The proposed CNN model is designed to automatically extract the optimal features for accurate disease classification, following a series of preprocessing techniques to enhance image quality. Our model demonstrated high accuracy and can make a good differentiation between healthy citrus fruits the diseased ones; hence this framework proved to be an esteemed tool for farmers and agriculturalists in the diagnosis of disease outbreaks which thus made our crop management practices improved ensuring sustainable production of citrus fruits.