Image Dataset for Disease Detection in Black Gram (Vigna mungo) Leaves: A Resource for Machine Learning Research

This dataset presents a curated collection of images of Black Gram (Vigna mungo) leaves, annotated with labels for healthy leaves and various common diseases. Created to support the advancement of machine learning and computer vision models in agricultural disease detection, this dataset is valuable for researchers and practitioners working in botany, plant pathology, agriculture, and artificial intelligence. The dataset is designed to reflect real-world agricultural conditions, providing a robust foundation for developing disease detection and classification models that can aid in crop health monitoring and management.

Dataset Content: The dataset includes original or raw data total of 4,038 images and augmented data total of 20,190 images (using rotation, brightness adjustment, horizontal flip, and zoom) representing healthy leaves and five distinct disease categories. Each category offers a range of visual variations, including different background conditions, lighting, and severity of disease symptoms, ensuring comprehensive data diversity. This resource can be used for training, testing, and validating machine learning models for image-based disease classification and detection tasks. The dataset is organized as follows:

Original data:
Healthy: 545 images
Cercospora leaf spot: 598 images
Leaf Crinkle: 806 images
Insect: 408 images
Yellow Mosaic: 1,681 images

Augmented data:
20,190 images

Purpose: The primary aim of this dataset is to facilitate the development of machine learning models that can accurately detect and classify diseases in Black Gram leaves, supporting early diagnosis and promoting effective crop management strategies. This dataset serves as a resource for improving automated plant disease diagnosis, contributing to agricultural sustainability and food security.

本数据集精选了黑豆（Vigna mungo）叶片的图像集，并对健康叶片及多种常见病害进行了标注。旨在支持机器学习和计算机视觉模型在农业病害检测领域的进步，该数据集对于植物学、植物病理学、农业及人工智能领域的科研人员和从业者而言具有重要价值。数据集旨在反映现实农业生产条件，为开发疾病检测和分类模型提供坚实的理论基础，此类模型有助于作物健康监测与管理。

数据集内容：该数据集包括4,038张原始或原始数据图像以及20,190张增强数据图像（通过旋转、亮度调整、水平翻转和缩放等手段生成），涵盖了健康叶片及五种不同的病害类别。每个类别均提供了丰富的视觉变体，包括不同的背景条件、光照和病害症状的严重程度，确保数据的全面多样性。本资源可用于训练、测试和验证基于图像的疾病分类和检测任务的机器学习模型。数据集的组织结构如下：

原始数据：
健康叶片：545张图像
叶斑病：598张图像
叶片皱缩：806张图像
昆虫：408张图像
黄色花叶：1,681张图像

增强数据：20,190张图像

目的：本数据集的首要目标是促进能够准确检测和分类黑豆叶片疾病的机器学习模型的发展，支持早期诊断并推动有效的作物管理策略。本数据集作为提高自动化植物病害诊断的资源，有助于促进农业可持续发展和粮食安全。