Smart Agriculture: AI-Driven Crop Disease Detection and Farmer Assistance

Abstract

Agricultural productivity together with loss reduc- tion depends heavily on prompt disease detection methods and disease classification procedures. The solution built around the InceptionV3 deep learning model provides identification of crop diseases based on uploaded images. Users register through the platform byverifyingtheir email withOTP thentheycanupload images that trigger disease prediction with precise classifications plus given treatment recommendations. Amultilingualchatbot poweredbyAIrunsthroughPerplexity API for instant communication about crop health and disease protection. The system provides complete data storage capabil- ities for predictions through its database structure so users can produce historical reports to improve their monitoring effec- tiveness and decision-making processes. The platform contains multilingual capabilities which allow diverse linguistic farmersto use its features. The project combines Django with SQLite3 database archi- tecture for building the backend alongside an interface frontend that employs HTML, CSS, JavaScript and Bootstrap program- ming elements. The system functions as a complete answer for agricultural operations where it helps farmers prevent diseasesin advance thus minimizing crop damage and creating better farming efficiency.

Introduction

The project develops an advanced crop disease detection system aimed at enhancing global food security by using deep learning and AI technologies. It leverages the InceptionV3 convolutional neural network to accurately identify 39 crop diseases from images uploaded by farmers. The system provides timely disease diagnoses, treatment recommendations, and pesticide guidelines to help prevent crop loss.

To improve accessibility, a multilingual chatbot powered by the Perplexity API offers real-time support and agricultural information in multiple languages, catering to farmers from diverse linguistic backgrounds. The platform also stores historical disease data to help farmers monitor disease trends and implement preventive measures.

The system is implemented with a Python Django backend and a responsive frontend using HTML, CSS, JavaScript, and Bootstrap. It integrates secure user authentication, image preprocessing, disease classification, chatbot interaction, and SQLite3 database management for storing user data, disease predictions, chatbot logs, and reports.

The solution combines AI-driven disease detection, multilingual user assistance, and long-term data tracking to provide an efficient, accessible, and accurate smart agriculture tool, ultimately supporting sustainable farming and reducing crop losses.

Conclusion

The development of a crop disease prediction system in- corporating deep learning and multilingual chatbot assistance has demonstrated its potential in supporting farmers and agricultural professionals. Thesystem successfullyutilizes the InceptionV3 model for disease classification, allowing usersto upload crop images and receive accurate disease diagnoses. The integration of the Perplexity API chatbot enhances ac- cessibility by offering instant guidance on crop diseases and preventive measures in multiple languages. Additionally, the historical report generation feature enables users to track dis- easeoccurrencesandmakedata-drivendecisionsforimproved crop health management. Themodel’sperformanceevaluationrevealedhighaccuracy in classifying crop diseases, making it a reliable tool for early disease detection. The chatbot integration provided real-time support, allowing farmers to receive quick responses to their agricultural queries. Furthermore, the system’s user-friendly webinterface, built with Djangoand Bootstrap, ensures seam- less interactionandeaseofusefor farmerswith varyinglevels of technical expertise. Despite its promising results, certain challenges were en- countered, such as similar disease patterns leading to misclas- sification and data imbalance for rare crop diseases. Addition- ally,real-timechatbot performancecan beenhanced furtherto improveresponseaccuracy.Nonetheless,thesystemprovidesa comprehensive,scalable,andefficientapproachtoagricultural disease detection, contributing to improved crop yield and sustainability.

References

[1] S. Mustofa, M. M. H. Munna, Y. R. Emon, G. Rabbany, and M. T.Ahad, “A comprehensive review on Plant Leaf Disease detection usingDeep learning,” arXiv preprint arXiv:2308.14087, 2023. [2] A. S. Abade, P. A. Ferreira, and F. de B. Vidal, “Plant Diseases recog-nition on images using Convolutional Neural Networks: A SystematicReview,” arXiv preprint arXiv:2009.04365, 2020. [3] A. S. Abade, P. A. Ferreira, and F. de B. Vidal, “Plant Diseases recog-nition on images using Convolutional Neural Networks: A SystematicReview,” arXiv preprint arXiv:2009.04365, 2020. [4] M.ElJarroudi,L.Kouadio,P.Delfosse,C.H.Bock,A.-K.Mahlein,X.Fettweis,B.Mercatoris,F.Adams,J.M.Lenne´,S.Hamdioui,and S.Mahlein ,“Enhancingplant disease detection throughdeeplearning,”FrontiersinPlantScience,vol.15,2024. [5] S. P. Mohanty,D. P. Hughes,and M. Salathe´,“Using deep learning forimage-basedplant diseasedetection,” Frontiersin PlantScience,vol.7, p.1419,2016. [6] P. Krishnan and A. K. Singh, “An Overview of Chatbots using MLAlgorithmsin AgriculturalDomain,” InternationalJournalofComputerApplications, vol. 184, no. 11, pp. 1–7, 2022. [7] M. Salathe´, C. L. Althaus, N. Anderegg, D. Antonioli, T. Ballouz, E.Bugnion,S.Capkun,D.Jackson,S.-I. Kim,J.Larus,N.Low,W.Lueks, [8] Menges, C. Moullet, M. Payer, J. Riou, T. Stadler, C. Troncoso, E.Vayena, and V. von Wyl, “Early Evidence of Effectiveness of DigitalContact Tracing for SARS-CoV-2 in Switzerland,” MedRxiv, 2020. [9] S. P. Mohanty,D. P. Hughes,and M. Salathe´,“Using deep learning forimage-basedplant diseasedetection,” Frontiersin PlantScience,vol.7, p.1419,2016. [10] S.Verma,“WhatisKissanAI?ExploreTheStartupStreamlining Farmers’ Lives!,” 2023. [Online]. Available:https://www.analyticsvidhya.com/blog/2023/12/what-is-kissanai-explore-the-startup-streamlining-farmers-lives/ [11] M. Salathe´, C. L. Althaus, N. Anderegg, D. Antonioli, T. Ballouz, E.Bugnion,S.Capkun,D.Jackson,S.-I.Kim,J.Larus,N.Low,W.Lueks, D.Menges,C.Moullet,M.Payer,J.Riou,T. [12] B.LalithaRajeswari,Sk.AbdulMuheeth,Sk.VaseemNaazleen, T. Pavan Kumar, V. Phanindraamouli (https://ijfans.org/./uploads/paper/e 8a57e632b18373eb83 97d4af4f58541.pdf)DOI:10.48047/IJFANS/V11/I12/179

Copyright

Copyright © 2025 Kiran Sai Nagendra Kappala, Lalitha Rajeswari Burra, Swapna Vunnam, Lakshmi Padmaja Patha , Md Sifa Minaaj . This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.