SAR-CLD-2024: A Comprehensive Dataset for Cotton Leaf Disease Detection

The Cotton Leaf Disease Detection Dataset represents a valuable resource for researchers, practitioners, and stakeholders in the agricultural sector, offering insights and tools to address the challenges associated with cotton leaf diseases effectively. Through accurate identification and classification of cotton leaf diseases, the dataset enables early detection, empowering farmers to take timely actions and optimize crop management strategies. Moreover, it supports the advancement of machine learning algorithms and methodologies for disease detection, fostering innovation in agricultural research. This dataset comprises meticulously curated images capturing various stages of cotton leaf diseases, sourced from the National Cotton Research Institute field in Gazipur. These images, captured using a Redmi Note 11s smartphone, represent a diverse range of disease manifestations across different dimensions. Despite challenges like fluctuating lighting conditions, field surveys conducted between October 2023 and January 2024, guided by domain experts, ensured high-quality image acquisition. The dataset contains 2137 images divided into seven classes, covering various cotton leaf conditions like bacterial blight, curl virus, and healthy leaves. These classes represent different issues affecting cotton plants, including diseases, pests, and environmental stress. Additionally, the dataset undergoes thorough preparation and augmentation procedures, including data cleaning, labeling, and augmentation techniques such as flipping and brightening. As a result, it comprises 2137 original images and 7000 augmented images, enhancing the effectiveness of deep learning models for precise classification and diagnosis of cotton leaf diseases. 1. Original (Cotton Leaf Disease Detection) Dataset: Number of datasets: 2137 Data format: .jpg 2. Augmented (Cotton Leaf Disease Detection) Dataset: Number of datasets: 7000 Data format: .jpg棉花叶片病害检测数据集（Cotton Leaf Disease Detection Dataset）是农业领域研究者、从业者与利益相关者的宝贵资源，可为有效应对棉花叶片病害相关挑战提供洞见与工具。通过精准识别与分类棉花叶片病害，该数据集可实现早期检测，助力农户及时采取应对措施、优化作物管理策略。此外，它还能推动病害检测相关机器学习算法与方法的迭代升级，助力农业研究领域的创新发展。 本数据集包含精心筛选的棉花叶片病害不同阶段图像，采集自加济布尔（Gazipur）国家棉花研究所试验田。这些图像由Redmi Note 11s智能手机拍摄，涵盖不同维度下多样化的病害表现形式。尽管存在光照条件波动等挑战，但2023年10月至2024年1月期间开展的野外调研在领域专家指导下完成，确保了图像采集的高质量。 该数据集共包含2137张图像，划分为7个类别，覆盖细菌性枯萎病（bacterial blight）、卷叶病毒病（curl virus）以及健康叶片等多种棉花叶片状态。这些类别涵盖了影响棉花植株的各类问题，包括病害、虫害与环境胁迫。 此外，该数据集经过了严格的预处理与增强流程，涵盖数据清洗、标注，以及翻转、亮度调整等数据增强技术。最终，数据集包含2137张原始图像与7000张增强图像，可提升深度学习模型对棉花叶片病害精准分类与诊断的效果。 1. 原始（棉花叶片病害检测）数据集：数据集数量：2137，数据格式：.jpg 2. 增强（棉花叶片病害检测）数据集：数据集数量：7000，数据格式：.jpg

搜集汇总

背景与挑战

背景概述

SAR-CLD-2024是一个专注于棉花叶部病害检测的综合性图像数据集，包含2137张原始图像和7000张增强图像，总计9137张.jpg格式图像，覆盖细菌性疫病、卷叶病毒病和健康叶片等七个类别。数据于2023年10月至2024年1月在孟加拉国加济普尔的棉花研究田间采集，使用智能手机拍摄并经过专家指导，确保在不同环境条件下的高质量图像。该数据集经过清理、标注和增强处理，旨在支持机器学习算法开发，帮助实现棉花病害的早期检测和分类，促进农业研究和作物管理优化。