SoyNet: A high-resolution Indian soybean image dataset for leaf disease classification.

Area of Science:

• Agricultural Science
• Computer Vision
• Machine Learning

Background:

• Accurate soybean disease identification is crucial for crop management.
• Existing datasets may lack the quality and diversity needed for robust model training.
• High-quality image data is essential for developing effective soybean disease classification systems.

Purpose of the Study:

• To introduce SoyNet, a novel, high-quality dataset for soybean leaf disease research.
• To provide a comprehensive resource for training, testing, and validating machine learning models for soybean disease classification.
• To address the need for diverse and realistic image data from soybean cultivation fields.

Main Methods:

• Curated over 9000 high-quality images of healthy and diseased soybean leaves.
• Captured images from various angles, lighting conditions, and backgrounds in agricultural fields.
• Organized data into raw and pre-processed (resized to 256x256, grayscale) folders.

Main Results:

• The SoyNet dataset offers a diverse collection of soybean leaf images.
• Includes both healthy and diseased leaf images, captured under real-world conditions.
• Provides pre-processed versions for streamlined model development.

Conclusions:

• The SoyNet dataset is a valuable resource for advancing soybean disease classification research.
• Facilitates the development and validation of more accurate and reliable AI models for agriculture.
• Supports improved disease management strategies through enhanced computational analysis.