Tea Tree Leaf and Manual Segmentation Dataset

The agricultural sector currently faces significant challenges in pest and disease monitoring and management, especially in tea tree cultivation. Traditional manual monitoring is inefficient and often lacks accuracy. Existing solutions typically rely on simple image processing technologies, which are insufficient in precision and adaptability. To address this issue, we have developed the Tea Tree Leaf and Manual Segmentation Dataset to improve the automatic recognition of leaf diseases. This dataset includes a rich collection of tea tree leaf images with corresponding manual segmentation annotations, designed to support smart agriculture. The data collection process involved capturing tea tree leaves in their natural environment using high-resolution cameras, with a systematic annotation process ensuring data quality. For quality control, we implemented multiple rounds of annotation and expert review to ensure consistency and accuracy. Data is stored in JPEG format and organized by image ID for easy retrieval and use. The dataset's strength lies in its high-quality annotations, achieving over 95% accuracy, and having undergone multiple rounds of validation to ensure consistency and completeness. Additionally, we utilized advanced image enhancement technologies to further improve data diversity and applicability. Models applied to this dataset have achieved a 15% increase in accuracy for pest and disease detection, providing more effective tools for precision agriculture.