Soil Samples and Remote Sensing Images Reference Dataset

The current agricultural sector faces challenges such as insufficient soil quality monitoring and difficulty in data acquisition. Traditional methods often rely on manual sampling, which is time-consuming and difficult to cover comprehensively. Although existing remote sensing technologies offer a broader view, they still have limitations in the accuracy of soil classification. This dataset aims to improve the accuracy of soil classification through high-quality soil samples and remote sensing image references. The dataset is collected by a professional team using high-resolution cameras to obtain remote sensing images under different plots and climatic conditions, and it is annotated. The data is stored in JPEG format, with a clear file organization structure, facilitating subsequent processing and analysis. To ensure data quality, multiple rounds of annotation and expert review are used to ensure consistency and accuracy of the annotations.