An India soyabean dataset for identification and classification of diseases using computer-vision algorithms

Area of Science:

• Agricultural Science
• Computer Vision
• Machine Learning

Background:

• Intelligent agriculture increasingly depends on accurate disease image recognition.
• Soybean is a significant crop in Maharashtra, India, grown across approximately 31,050 hectares.
• Effective disease identification is crucial for maintaining crop yield and quality.

Purpose of the Study:

• To create and release a comprehensive dataset of healthy and diseased soybean plant leaf images.
• To support research in agricultural disease image recognition and machine learning model development.
• To provide researchers and students with accessible data for academic and practical applications.

Main Methods:

• Collected 3363 images of soybean plants over two to three seasons from multiple farms.
• Categorized images into six classes: Healthy plants, Vein Necrosis, Dry leaf, Septoria brown spot, Root images, and Bacterial leaf blight.
• Organized the dataset into seven distinct folders for ease of access.

Main Results:

• A diverse dataset of 3363 images featuring various soybean plant conditions was established.
• The dataset includes images representing healthy plants and specific diseases like Vein Necrosis, Dry leaf, Septoria brown spot, and Bacterial leaf blight.
• Root images were also included, offering a broader scope for analysis.

Conclusions:

• The released dataset is a valuable resource for advancing agricultural disease image recognition.
• It facilitates the development and benchmarking of machine learning models for intelligent agriculture.
• Accessibility of this dataset will foster further research and innovation in crop health management.