The active region model (ARM) parameter (γ) controls the generation and development of the preferential flow. Therefore, to obtain an accurate ARM parameter (γ) is essential for getting a precise prediction using ARM. In this study, dye staining methods were used to visualize the preferential flow patterns from their flow background, and the digital imaging procedure and soil sampling were conducted to obtain the distributions of dye stained region and soil water content of stained region, from which the ARM parameter (γ) was estimated by fitting these data according to the constitutive relation of ARM after infiltration. The determination methods for ARM parameter (γ) were presented according to 3 different kinds of distribution patterns of the soil water content of stained region after infiltration. The results indicated that, 1) as a result of soil water redistribution, the active region and the dye stained region might not coincide with each other in the entire infiltrated soil profile, therefore, not all the data (soil water content and fraction of the dye stained region) from the entire infiltrated soil profile but only the data from the soil layer where the soil water contents were under increasing and less affected by the initial soil water contents could be used to determine the ARM parameter (γ); 2) the soil texture generally had obviously significant impacts on the distribution of soil water content after infiltration, making the maximum soil water content of stained region occurred more probably at the soil surface in the fine-textured soil (e.g. silty clay and loam in this research) while making the maximum soil water content of stained region occurred more probably under the soil surface in the coarse-textured soil (e.g. sand in this research).