# Detection of healthy and diseased crops in drone captured images using Deep Learning

**Authors:** Jai Vardhan, Kothapalli Sai Swetha  
**arXiv:** [2305.13490](https://arxiv.org/abs/2305.13490)  
**Date:** 2023-05-22  
**Summary page:** https://www.summarizepaper.com/en/paper/2305.13490v1/

## Key points

- Deep learning-based approach for efficient detection of plant diseases using drone-captured imagery
- Importance of monitoring plant health for maintaining agricultural productivity and food safety
- Timely detection of diseases crucial to prevent crop loss
- Compilation of comprehensive database of various plant species exhibiting numerous diseases from the Internet
- Use of Convolutional Neural Network (CNN) as primary predictive model for crop disease categorization and detection
- Deployment of prototype drone model equipped with high-resolution camera for live monitoring of extensive agricultural fields
- Real-time identification of healthy and diseased plants through trained model using captured images
- Promises efficient and scalable solution for improving crop health monitoring systems
- Novel application of deep learning techniques in agriculture leveraging drone-captured imagery
- Potential to significantly reduce crop loss and improve agricultural productivity while ensuring food safety

## Summary

In this study, authors Jai Vardhan and Kothapalli Sai Swetha propose a deep learning-based approach for efficient detection of plant diseases using drone-captured imagery. They emphasize the importance of monitoring plant health for maintaining agricultural productivity and food safety as disruptions in the plant's normal state caused by diseases can significantly impact essential plant activities and crop yield. Timely detection of these diseases is crucial to prevent crop loss. To develop their approach, the authors compiled a comprehensive database of various plant species exhibiting numerous diseases from the Internet. This dataset was used as the training and test dataset for their model. They employed a Convolutional Neural Network (CNN), known for its performance in image classification tasks, as their primary predictive model. The CNN model trained on this rich dataset demonstrated superior proficiency in crop disease categorization and detection even under challenging imaging conditions. For field implementation, the authors deployed a prototype drone model equipped with a high-resolution camera to enable live monitoring of extensive agricultural fields. The captured images served as input for their trained model, enabling real-time identification of healthy and diseased plants. This approach promises an efficient and scalable solution for improving crop health monitoring systems. Overall, this study presents a novel application of deep learning techniques in agriculture by leveraging drone-captured imagery to detect plant diseases. By providing timely disease detection, this approach has the potential to significantly reduce crop loss and improve agricultural productivity while ensuring food safety.