Multi-component gradient enhancement for accurate frost detection and quantification on leaf surfaces.

Area of Science:

• Agricultural Science
• Computer Vision
• Image Processing

Background:

• Accurate leaf frost detection is vital for agricultural monitoring.
• Existing methods face segmentation challenges due to complex backgrounds and subtle frost-leaf textures.

Purpose of the Study:

• To develop an advanced method for precise leaf frost detection.
• To overcome limitations of current techniques in complex environments.

Main Methods:

• Proposed MCGE-Frost (Multi-Component Gradient Enhancement) method.
• Integrates color space analysis (HSV, Lab) with gradient fusion theory.
• Employs adaptive weighting and morphological filtering for enhanced contrast and noise reduction.

Main Results:

• MCGE-Frost achieved a 3.29% segmentation error rate, outperforming ExG (8.63%), OTSU (8.98%), and HSV (11.98%).
• Reduced computational complexity by 40% compared to deep learning methods.
• Achieved 0.8 s/image processing on GPU-accelerated systems.

Conclusions:

• MCGE-Frost offers a robust and efficient solution for real-time leaf frost detection.
• The method enhances frost quantification intelligence for precision agriculture.
• Supports actionable insights for frost protection and crop management.