Quantitative Study on Salinity Estimation of Salt-Affected Soils by Combining Different Types of Crack Characteristics Using Ground-Based Remote Sensing Observation

Soil salinity is one of the parameters used for determining extent soil salinization. During water evaporation, surface salt-affected soils in Songnen Plain, China, exhibits obvious shrinkage and cracking phenomena due to high salt content. The aim this current study quantify influence content on shrinkage–cracking process achieve quantitative extraction based different crack parameter types. In order above objectives, a controlled experiment was conducted. Subsequently, three kinds characteristics such as length, box-counting dimension, 12 gray-level co-occurrence matrix (GLCM) texture features were quantitatively extracted from standard binary patterns. predict physical–chemical properties models multiple linear regression (MLR), stepwise (MSR), artificial neural network (ANN) developed compared first two principal components GLCM features. results show that desiccation cracks determined by since film caused exchangeable cations thickness DDL can promote cracking. Although methods have prediction accuracy Na+, electrical conductivity (EC), total salinity, ANN-based method showed best with R2 values EC, 0.91, 0.89, ratio performance deviation (RPD) corresponding 2.96, 3.47, 2.95.