Pomegranate disease detection and classification dataset for deep learning applications: A case study from Halabja...

Area of Science:

• Agricultural Science
• Computer Vision
• Machine Learning

Background:

• Accurate detection of pomegranate fruit diseases is vital for agricultural sustainability and economic viability.
• Existing datasets may lack the diversity and real-world conditions needed for robust AI model training.
• Pomegranate production faces significant losses due to diseases and environmental factors like sunburn.

Purpose of the Study:

• To introduce the Halabja Pomegranate Fruit Disease Image Dataset for research and development.
• To provide a diverse and contextually relevant image collection for training disease detection models.
• To facilitate the creation of effective diagnostic tools for real-world field conditions.

Main Methods:

• Compilation of 2178 original and 28,314 augmented images from Iraqi pomegranate orchards.
• Categorization into four classes: ectomyelois ceratoniae, colletotrichum spp., sunburn, and healthy fruit.
• Image preprocessing included resizing to 512x512 pixels and applying augmentation techniques.

Main Results:

• The dataset offers significant class variation captured in natural outdoor environments.
• Preprocessing enhances model flexibility and robustness for machine learning applications.
• The dataset's unique characteristics are suitable for developing precision agriculture computer vision tools.

Conclusions:

• The Halabja Pomegranate Fruit Disease Image Dataset is a valuable resource for AI-driven plant disease detection.
• This dataset can support the development of diagnostic tools for real-world agricultural challenges.
• It contributes to advancing computer vision applications in precision agriculture and crop management.