Soil color visually reflects soil conditions and is thus widely employed to evaluate relevant classifications and characteristics. However, predicting soil conditions based on soil color can be unreliable due to the influence of lighting environments. This study aimed to predict soil components using digital images taken under variable lighting conditions. The correlation between CIELAB-based soil color and lighting conditions was analyzed, and intercept values from the linear regression equation between CIELAB-based soil color and lighting conditions were identified as reliable indicators of soil components and water content, regardless of lighting conditions. A soil color-based model for predicting soil components and water content was developed using 270 soil images and validated with an additional 42 soil images. The developed model provides a rapid, objective, and accurate assessment of soil components and water content under varying lighting conditions, thereby enhancing the applicability of soil color-based predictions in field scenarios.
Discussion(0)
No comments yet. Be the first to comment.