LuffaFolio: A Multifaceted Luffa Aegyptiaca Image Dataset

Published: 12 December 2023| Version 3 | DOI: 10.17632/kwj9599z73.3


This dataset is divided into 3 parts; Each part consists of smooth luffa grading, diseases, and flowers respectively. These images were captured from different village fields of Faridpur, Bangladesh. To construct the dataset, we have considered Luffa Aegyptiaca commonly known as Smooth Luffa (Dhundal/ধুন্দল). This dataset consists of a total of 1,933 JPG images; the dimension of each image is 1728 x 1728 pixels. The total size of the dataset is 1.45 GB. The dataset contains 3 folders: Luffa_Diseases, Flowers, and Luffa_Grade. The detail of the dataset: • Luffa_Diseases: This dataset contains leaves of smooth luffa representing various diseases along with non-affected ones. The categories of this dataset are Alternaria Disease, Angular Spot Disease, Holed Leaves, Mosaic Virus, and Fresh Leaves. A total of 1,228 JPG raw images are presented in this folder. • Flowers: This dataset contains flowers of smooth luffa. There is only a single category of this dataset. A total of 362 JPG raw images are presented in this folder. These images represent various maturity stages of smooth luffa flowers. • Luffa_Grade: This dataset contains fresh and defective smooth luffa. The categories of this dataset are Fresh and Faulty Luffa. A total of 343 JPG raw images are presented in this folder. We have also uploaded a Resized version (224 X 224 pixels) of full dataset in zip format.



American International University Bangladesh, Jahangirnagar University, Bangladesh University of Business and Technology


Artificial Intelligence, Computer Vision, Image Processing, Machine Learning, Computer Imaging, Agricultural Development, Deep Learning