Answers are grounded in this paper and cite their sources.
Arecanut is an important plantation crop in Karnataka, and delayed recognition of plant-health problems can reduce yield and farm income.Manual inspection is subjective and time-consuming, and technical crop-health information is rarely available in the regional language.This paper presents a preliminary decision-support system that couples deep-learning image classification with a Kannada-language chatbot interface for arecanut farmers.An uploaded image is resized to 227 × 227 pixels, rescaled, and classified by a sequential convolutional neural network (CNN) with an AlexNet-style arrangement of five convolutional layers and batch normalization, trained with the Adam optimizer and rotation, zoom and horizontal-flip augmentation for 50 epochs.A Flask web application returns the predicted class with related symptom and management information and hosts the "Areca Assistant" chatbot, which answers follow-up questions in Kannada.After 50 epochs the model reached a training accuracy of 99.03% and a validation accuracy of 99.67%, and a scikit-learn classification report on 8,143 evaluation samples gave macro-averaged precision, recall and F1-score of 0.98, 0.99 and 0.98.The labelled data span several crops, and an independent test set, field trials and a quantitative chatbot evaluation remain to be carried out.The main contribution is the integration of image-based screening and regional-language assistance in a single workflow that can support smallholder arecanut growers in early disease awareness.
No takes yet. Share an insight, caveat, or question.
Santosh et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: