AI-Based Pest and Disease Detection in Crops

  • Unique Paper ID: 177145
  • Volume: 11
  • Issue: 12
  • PageNo: 64-69
  • Abstract:
  • The increasing global demand for food necessitates enhanced crop productivity, which is critically hindered by plant diseases and pest infestations. Conventional detection methods are predominantly manual, time-consuming, and ineffective for large-scale agricultural monitoring. This paper presents an integrated, AI-based approach for automated detection of crop diseases and pests using deep learning techniques. For disease identification, a custom Convolutional Neural Network (CNN) is trained on the disease detection dataset to classify potato leaf conditions into three categories: early blight, late blight, and healthy. The model incorporates data augmentation and image preprocessing to enhance generalization, achieving a test accuracy of 96.3%. For pest detection, a transfer learning framework utilizing the MobileNet architecture is employed. The model is fine-tuned on a dataset comprising nine pest classes and is optimized using advanced image augmentation techniques. The resulting classifier demonstrates a test accuracy of 96.22%, indicating high reliability and scalability in field conditions. The proposed dual-model framework offers a non-invasive, high-throughput solution for real-time monitoring of crop health. This work contributes to the development of intelligent precision agriculture systems by supporting early detection, timely intervention, and informed decision-making.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 11
  • Issue: 12
  • PageNo: 64-69

AI-Based Pest and Disease Detection in Crops

Related Articles