Identification of key factors for soil drainage classification via color metric comparison from soil images

다국어 초록 (Multilingual Abstract)

Drainage class is a critical factor in evaluating soil for crop suitability, water management, and sustainable agricultural planning. Traditional classification methods, which rely on field surveys and expert judgment, can be subjective. This study aims to establish objective criteria for soil drainage classification by analyzing color properties extracted from soil images. A total of 88 soil samples were collected from diverse parent materials and drainage classes. Soil colors were quantified using the Munsell soil color system, HSV, and CIELAB color spaces. In well drained and moderately well drained soils, matrix colors exhibited high chroma (≥3), while redoximorphic features showed low chroma (1 - 2), indicating strong oxidizing conditions. In contrast, poorly drained soils displayed grayish matrix colors with low chroma (≤2) and reddish mottles with high chroma (≥4), reflecting alternating redox conditions. Hue values in the HSV space and a* (red - green axis of the CIELAB color space) values in the CIELAB space were particularly effective in distinguishing drainage classes: well drained soils generally had higher Hue and a* values, whereas poorly drained soils tended toward lower Hue and negative a* values. Phyllite derived soils exhibited unique characteristics due to the inherent color of the parent material, highlighting the need for integrated interpretation. These findings demonstrate that digital image based soil color analysis, particularly using Hue and a* values, provides a quantitative and scalable approach to classifying soil drainage classes.