Using Deep Learning for Image-Based Potato Tuber Disease Detection.

摘要: Many plant diseases have distinct visual symptoms, which can be used to identify and classify them correctly. This article presents a potato disease classification algorithm that leverages these appearances advances in computer vision made possible by deep learning. The uses convolutional neural network, training it the tubers into five classes: namely, four classes healthy class. database of images this study, containing different cultivars, sizes, diseases, was acquired, classified, labeled manually experts. models were trained over train-test splits better understand amount image data needed apply learning for such tasks. tested set taken using standard low-cost RGB (red, green, blue) sensors tagged experts, demonstrating high accuracy. is first report successful implementation networks, popular object identification, task identification tubers, showing potential techniques agricultural