LuffaFolio: A Multidimensional Image Dataset of Smooth Luffa

Area of Science:

• Agricultural Science
• Plant Pathology
• Computer Vision

Background:

• Accurate classification of Luffa diseases and quality is crucial for agricultural productivity.
• Existing datasets may lack comprehensive coverage of Luffa Aegyptiaca (Smooth Luffa) across various aspects like disease, grade, and growth stages.
• Image-based analysis offers a scalable approach for agricultural monitoring.

Purpose of the Study:

• To introduce a comprehensive, multi-faceted image dataset for Luffa Aegyptiaca.
• To facilitate research in Luffa disease classification, quality grading, and growth stage identification.
• To support the development of automated systems for Luffa crop management.

Main Methods:

• A dataset of 1933 JPG images was collected from village fields in Faridpur, Bangladesh.
• The dataset is organized into three sections: Luffa Diseases (1228 images), Flowers (362 images), and Luffa Grade (343 images).
• Images capture various Luffa diseases (e.g., Alternaria, Angular Spot, Mosaic Virus), flower maturity stages, and Luffa quality (fresh vs. defective).

Main Results:

• The dataset provides a diverse collection of Luffa images categorized for specific agricultural research tasks.
• It includes detailed classifications for common Luffa leaf diseases and distinct growth stages of flowers.
• The Luffa Grade section offers images for distinguishing between fresh and defective produce.

Conclusions:

• This dataset serves as a valuable resource for advancing research in Luffa Aegyptiaca cultivation and management.
• It enables the development and validation of machine learning models for automated Luffa analysis.
• The dataset's comprehensive nature supports diverse applications in plant pathology and agricultural quality assessment.