SavingNeem: An image dataset for Neem leaf disease classification

To address the challenges in automated detection and classification of Neem (Azadirachta indica) leaf diseases, having access to a comprehensive and high-quality annotated image dataset is essential. This work presents a carefully curated dataset called "Neem Leaf Disease" consisting of 12880 colour images. The images were captured using mobile phone cameras under various angles to represent real-world diversity. And the leaves were collected from different areas of Dhaka and nurseries across multiple areas. The dataset is organised in a total of 14 folders. At first all the pictures were resized as 640p X 480p and then these images were augmented. Images are sorted into separate folders for each disease class and identification. With its diverse images and organisation, this dataset can facilitate training, evaluation, and benchmarking of algorithms aimed at automated monitoring of Neem plant health for precision agriculture and crop management applications.