Ai Meets Nature: Crop Disease Detection With Organic Remedy Recommendations

• Author(s):

  Gorisetty Nirosha | Sruthilaya Mindi | Gembali Sravanthi | Ungati Mahesh | Shaik Afrin

• Keywords:

  Vision Transformer (ViT), Crop Disease Detection, Organic And Indigenous Remedies, Segmentation, Soil Health Restoration.

• Abstract:

  Food Security And Sustainable Agriculture Are Critical Global Priorities, Yet Current Farming Practices Heavily Reliant On Chemical Pesticides Continue To Degrade Soil Health, Reduce Biodiversity, And Pose Risks To Human Well-being. Most Existing AI-based Crop Disease Detection Systems Focus Solely On Classification And, When Treatment Suggestions Are Provided, They Predominantly Recommend Synthetic Chemicals. Such Approaches Address The Symptom But Further Harm Long-term Soil Fertility And Ecosystem Balance. This Project Presents An Intelligent, Eco-friendly, Web-based AI Platform That Not Only Detects Crop Diseases With High Accuracy But Also Recommends Sustainable, Organic Treatments. Advancing Beyond Conventional Convolutional Neural Network (CNN) Models Used In Prior Work, The Proposed System Leverages Vision Transformers (ViT) For Superior Global Feature Extraction, Enabling Robust Detection Of Subtle Disease Patterns Even Under Challenging Image Conditions. To Enhance Interpretability And Facilitate Precise Localization Of Infected Areas, ViT-based Segmentation Techniques Are Integrated, Allowing Clear Visual Mapping Of Diseased Regions For Better Understanding And Validation. In Addition To Detection, The Platform Incorporates A Curated Knowledge Base Of Organic And Indigenous Remedies, Rooted In Traditional Farming Wisdom And Validated Through Modern Agronomic Research. This Integration Significantly Reduces Dependence On Harmful Chemicals, Promoting Cost-effective, Environmentally Responsible Practices That Enhance Soil Fertility And Crop Resilience.

---

Other Details

• Paper id:

  IJSARTV11I11104247

• Published in:

  Volume: 11 Issue: 11 November 2025

• Publication Date:

  2025-11-06

Download Article