PLANT DISEASE DETECTION AND FERTILIIZER RECOMMENDATION USING ML AND DL

Abstract

Agriculture remains a cornerstone of the global economy and food security, but it faces major challenges due to crop diseases and imbalanced fertilizer usage. These issues can lead to significant yield losses and long-term soil degradation. Early detection of plant diseases and the application of appropriate fertilizers are vital to improving agricultural productivity. This paper proposes a dual-purpose system leveraging machine learning (ML) and deep learning (DL) to assist farmers in identifying plant diseases and recommending suitable fertilizers.
The plant disease detection module uses Convolutional Neural Networks (CNNs), a class of deep learning models known for their high performance in image classification tasks. Leaf images are processed and classified into healthy or diseased categories, with further identification of specific disease types. This approach reduces the need for manual inspection and expert intervention, providing farmers with instant diagnostic capabilities. Transfer learning with pre-trained models such as VGG16 or ResNet is applied to enhance accuracy and reduce training time.