locations mentioned below. Location 1: Bu Dang District, Binh Phuoc Province, Vietnam Location 2: Cai Lay District, Tien Giang Province, Vietnam Data accessibility Repository: A Durian Leaf Image Dataset of Common Diseases in Vietnam for Agricultural Diagnosis Data identification number: 10.17632/pxzvksbwnj Direct URL to data: https://data.mendeley.com/datasets/pxzvksbwnj Instructions for accessing these data: The data is divided into three sets: train, test, and validation. Each set contains six folders with pre-processed images in JPG format. Related research article 1. Value of the Data • The dataset consists of 2595 images of durian leaves captured via mobile devices, including healthy l
Open resource ↗10.17632/pxzvksbwnj · lines:1-54Unverified paper record
A durian leaf image dataset of common diseases in Vietnam for agricultural diagnosis.
Data in brief · 30 Jun 2025 · 10.1016/j.dib.2025.111845
Abstract
Agriculture plays a vital role in Vietnam's economy, with durian being a key high-value crop that supports millions of farmers. However, durian leaves are highly susceptible to pests, diseases, and environmental stressors, negatively impacting yield and quality. This study introduces a dataset of 2595 durian leaf images, categorized into six classes: 484 healthy leaves and 2111 diseased leaves spanning Blight (440), Colletotrichum (400), Algal (462), Phomopsis (411), and Rhizoctonia (398). The images were collected from durian orchards across Vietnam under diverse conditions, then background-removed, resized to 400 × 400 pixels, and manually annotated with expert guidance. This dataset provides a valuable resource for advancing research in automated plant disease detection, enabling the development of computer vision models for early diagnosis and precision farming, thereby supporting sustainable durian production and improved crop productivity.
Plant phenotyping relevance
罹病葉の画像を収集・前処理・専門家注釈したデータセットの構築が中心で、植物病害状態の画像ベース表現型判定に直接利用できる。
abstractThis study introduces a dataset of 2595 durian leaf images, categorized into six classes
abstractThe images were collected from durian orchards across Vietnam under diverse conditions, then background-removed, resized to 400 × 400 pixels, and manually annotated with expert guidance.
abstractThis dataset provides a valuable resource for advancing research in automated plant disease detection
Code and data availability
The paper is a data descriptor for a durian leaf disease image dataset (2595 annotated images) publicly deposited on Mendeley Data with DOI 10.17632/pxzvksbwnj and direct URL provided in the text.
This is an automatically classified, unverified record. Curator approval is required before any resource enters the Catalog.