EarlyNSD

The EarlyNSD Dataset is a pioneering resource for the early detection of nutrient deficiencies in plants, focusing on cucurbit crops such as ash gourd (Benincasa hispida), bitter gourd (Momordica charantia), and snake gourd (Trichosanthes cucumerina). With over 2,700 segmented and augmented leaf images, this dataset is specifically designed to identify early signs of nitrogen and potassium deficiencies that can significantly impact crop yield.

Key Features:

• Targeted Nutrient Deficiencies: Focuses on early indicators of nitrogen and potassium deficiencies in cucurbit leaves, essential for sustainable farming practices.
• Three Important Cucurbits: Includes images of ash gourd, bitter gourd, and snake gourd, vegetables that are crucial to global food production.
• Large and Diverse Dataset: 2,700 segmented leaf images, enriched through data augmentation to increase variability and enhance model training.
• Proven Impact on Crop Yield: Designed to assist in the early diagnosis of plant stress, contributing to improved crop management and yield preservation.

Dataset Structure:

• Training: 70% of the dataset for model training.
• Validation: 15% for evaluating model performance during training.
• Testing: 15% for final model evaluation.

Applications:

• Agricultural Research: Focuses on identifying early nutritional stress in cucurbit crops, a crucial step for maximizing crop yield and reducing the impact of nutrient deficiencies.
• Crop Health Monitoring Systems: Ideal for developing diagnostic tools that help farmers monitor plant health in real-time.
• AI and Machine Learning in Agriculture: Enables training of machine learning models to automatically detect early signs of nutrient stress, improving precision agriculture techniques