Fruits and Vegetables Disease Detection Using Deep Learning

Fruits and vegetables play a crucial role in maintaining overall health, offering essential nutrients and numerous health benefits. Advancements in computer vision technology have simplified various agriculture-related issues, including the detection of fruit and vegetable diseases. In this research, we introduce a deep learning-driven system for classifying the disease of grapes, guavas, pomegranates, cucumbers, potatoes, and tomatoes as healthy or rotten. ResNet50 achieved the highest accuracy at 98%, with MobileNetV2 following closely at 89%, Xception at 66%, and InceptionV3 at 62%. Our methodology involves comprehensive image acquisition, pre-processing, and segmentation techniques to enhance feature extraction. This system enables farmers to effectively identify and separate healthy produce from rotten ones, ensuring better crop management and reduced waste. Our findings highlight the potential of CNNs in agricultural applications, providing a robust solution for maintaining crop quality and supporting sustainable farming practices. Future research directions are also discussed to further refine these techniques.