Abstract:
Tomato plants have a high specificity of different diseases, which reduce crop yield and fruit
quality, directly reducing agricultural productivity. Most current solutions are restricted to disease
detection alone and do not provide an integrated system to detect plant diseases with suggestions
for remedies. Moreover, these models are not user-friendly and do not represent environmental
heterogeneity, therefore, their applicability is constrained for farmers in different environmental
conditions.
AgroAI is a machine learning-based algorithm that utilizes Convolutional Neural Networks (CNN)
for crop disease detection and provides solutions based on disease severity. Using a curated dataset
of tomato leaf images for diseases like Target Spot, Late Blight, and Bacterial Spot, the system
processes images uploaded by farmers, classifying diseases. Data augmentation and pre
processing techniques were applied to enhance model robustness, and a straightforward interface
was developed to facilitate ease of use for farmers.
