Related Experiment Video
Updated: Mar 17, 2026
08:47
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
Published on: February 9, 2024
2.1K
Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification
Srdjan Sladojevic1, Marko Arsenovic1, Andras Anderla1
1Department of Industrial Engineering and Management, Faculty of Technical Sciences, University of Novi Sad, Trg Dositeja Obradovica 6, 21000 Novi Sad, Serbia.
Computational Intelligence and Neuroscience
|July 16, 2016
Summary
This study introduces a novel deep convolutional network approach for plant disease recognition using leaf images. The developed model accurately identifies 13 diseases, offering a practical solution for agriculture.
Area of Science:
- Agricultural Science
- Computer Science
- Machine Learning
Background:
- Convolutional Neural Networks (CNNs) have shown significant success in image classification tasks.
- Accurate plant disease recognition is crucial for effective agricultural management and food security.
- Existing methods may lack the speed and accuracy required for practical, large-scale implementation.
Purpose of the Study:
- To develop and present a novel deep convolutional network-based model for plant disease recognition.
- To enable the classification of 13 distinct plant diseases from leaf images.
- To create a system capable of distinguishing plant leaves from their background.
Main Methods:
- Utilized deep convolutional networks for leaf image classification.
- Developed a novel training methodology for efficient system implementation.
- Employed the Caffe deep learning framework for model training.
- Created a comprehensive image database assessed by agricultural experts.
Main Results:
- The model achieved high precision, ranging from 91% to 98% for individual class tests.
- An average precision of 96.3% was recorded across all tested classes.
- The system demonstrated the ability to accurately differentiate healthy leaves from diseased ones and segment leaves from surroundings.
Conclusions:
- The proposed deep convolutional network approach represents a novel and effective method for plant disease recognition.
- The developed model offers a practical, accurate, and efficient solution for identifying multiple plant diseases.
- This research lays the groundwork for implementing advanced AI-driven disease detection systems in agriculture.
