A survey: to identify plant leaf diseases by feature extraction methods and classification techniques

Area of Science:

• Agricultural Science
• Computer Science
• Artificial Intelligence

Background:

• Deep learning (DL) methods are revolutionizing image-based plant disease diagnosis.
• Automated plant disease identification enhances crop monitoring and agricultural productivity.
• Accurate disease identification is crucial for effective crop management and sustainable agriculture.

Purpose of the Study:

• To provide an extensive overview of Convolutional Neural Networks (CNNs) for plant disease detection.
• To highlight recent advancements in CNN-based models for detecting plant leaf diseases.
• To analyze innovations, methods, and challenges in applying CNNs for plant health monitoring.

Main Methods:

• The study surveys recent research (last half-decade) on CNN-based deep learning models.
• Focuses on deep convolutional neural networks (DCNNs) trained on large-scale image databases.
• Examines the application of CNNs for early and precise detection of plant diseases.

Main Results:

• CNN-based DL models demonstrate significant potential for early and accurate plant disease detection.
• Deep convolutional neural networks are proving effective in identifying plant diseases from images.
• Recent research highlights innovations and effective methods in CNN-based plant disease diagnosis.

Conclusions:

• CNN-based deep learning offers opportunities for early and accurate plant disease detection, supporting sustainable agriculture.
• Potential challenges exist in the practical, real-world application of these models.
• Future directions for DL-aided plant disease diagnosis are explored, with a critique of CNNs' potential and limitations.