Drone-Based Cotton Plant Health Dataset: Healthy vs. Unhealthy Classification

This dataset comprises high-resolution drone-based aerial imagery of cotton fields with corresponding annotations for plant health classification. The dataset contains RGB images captured using Unmanned Aerial Vehicles (UAVs) at low altitude, enabling detailed observation of individual cotton plants and their health status. Each image is accompanied by bounding box annotations in YOLO format, identifying and classifying cotton plants as either healthy or unhealthy based on visual characteristics such as leaf color, canopy density, wilting, discoloration, and disease symptoms.The dataset includes 300+ frames extracted from continuous drone video footage, with pixel-level annotations for precise localization of plants. The labeling scheme uses a binary classification system: class 0 represents plants exhibiting visual stress indicators (yellowing, necrosis, wilting, disease symptoms), while healthy plants are labeled accordingly. This dataset is designed to support the development and training of machine learning and deep learning models for automated cotton plant health monitoring, disease detection, and precision agriculture applications.The high spatial resolution of the aerial imagery and the detailed bounding box annotations make this dataset valuable for computer vision tasks including object detection, image segmentation, and crop health assessment. This resource can facilitate research in automated crop monitoring, early disease detection, yield prediction, and optimization of agricultural management practices through remote sensing and artificial intelligence technologies.Dataset Characteristics:Image format: RGB JPGAnnotation format: YOLO (normalized bounding boxes)Classification: Binary (Healthy\/Unhealthy)Data volume: 600+ frames with corresponding labelsSpatial resolution: High-resolution aerial imageryApplication domain: Precision agriculture, crop health monitoring, disease detection