Article Details
  • Published Online:
    April  2026
  • Product Name:
    The IUP Journal of Computer Sciences
  • Product Type:
    Article
  • Product Code:
    IJCS040326
  • DOI:
    10.71329/IUPJCS/2026.20.2.28-43
  • Author Name:
    Rishitosh Mondal, Prarambhika Bhattacharjee, Piyush Dawn, Koushiki Ghosh, Debolina Saha and Sudip Dogra
  • Availability:
    YES
  • Subject/Domain:
    Engineering
  • Download Format:
    PDF
  • Pages:
    28-43
Volume 25, Issue 2, April-June 2026
A Farmer-Friendly Mobile Framework for Real-Time Pest and Disease Detection and Remedial Guidance Using Deep Learning with Multilingual and Offline Support
Abstract

Crop pests and plant diseases significantly threaten agricultural productivity, particularly in rural regions with limited access to expert support and reliable internet connectivity. This paper presents CropGuard—a farmer-centric mobile decision support system—for realtime pest and disease detection. The proposed framework utilizes a YOLOv8-based object detection model deployed on-device to enable low-latency and offline image analysis. Detected diseases are linked to an expert-curated recommendation engine derived from official government agricultural portals, providing validated preventive and treatment guidance through a simplified multilingual interface supporting five languages. To enhance field-level usability, the system incorporates offline result storage and notification delivery. Experimental evaluation demonstrates reliable detection performance suitable for realworld agricultural deployment, effectively bridging the gap between advanced computer vision techniques and accessible, practical farming solutions.

Introduction

Crop pests and plant diseases continue to pose serious challenges to agricultural productivity, particularly in regions where access to timely expert guidance is limited.