Soil color calibration of digital images using a reference color patch

Soil color is a key indicator in soil science, used to assess soil properties such as mineral composition, organic matter content, and moisture content. However, soil color is highly dependent on lighting conditions, which vary significantly in real-world environments. To address this challenge, this study proposes a real-time and cost-effective soil color calibration method that integrates lighting condition measurement and image acquisition into a single step. Digital images of 24 industry-standard color patches were captured under irregular lighting conditions, and the relationship between patch color and lighting conditions was analyzed using the CIELAB color system. Among these, the brown color patch, which closely resembles soil color, was identified as the optimal reference color patch, enabling the development of empirical equations to estimate illuminance and color temperature. These equations allowed real-time calibration of soil color, standardizing soil images taken under different lighting conditions to a desired lighting condition. The proposed method was validated using eighteen digital images of moist soil captured under natural lighting conditions, demonstrating high calibration accuracy. This approach offers a practical, low-cost, and computationally efficient alternative to spectrophotometer-based methods, making real-time soil color calibration feasible for field applications.