中文摘要：

为有效和无损地监测土壤表层含水率,以壤土、红壤和黑壤为研究对象,研究了不同密度下土壤含水率（θ）的图像信息,包括亮度（V）和饱和度（S）,建立了土壤含水率的估算模型。结果表明,①当壤土密度为1 400 kg/m^3时,估算模型相关性最好（决定系数（R^2）为0.93,均方根误差（RMSE）是0.04 m^3/m^3）;②当红壤密度为1 600 kg/m^3时,相关性最好（R^2为0.93,RMSE为0.03 m^3/m^3）;③黑壤密度为1 600 kg/m^3时,相关性最好（R^2为0.87,RMSE为0.07 m^3/m^3）。考虑密度变化时,黏壤土的估算精度最好（R^2为0.72和RMSE为0.06 m^3/m^3）、其次是壤土（R^2和RMSE为0.70和0.07 m^3/m^3）和砂质壤土（R^2为0.60和RMSE为0.07 m^3/m^3）。因此,利用数码相机来估测表层不同密度土壤含水率是可行的。

英文摘要：

Taking loam, red and black soils as examples, we investigated the impact of bulk density on accuracy of the soil water content estimated using digital images such as their brightness and saturation. The results show that for the less dese loam at bulk density of 1 400 kg/m^3, the correlation coefficient（R^2） between image information and water content is 0.93 with a root mean square error（RMSE） of 0.04 m^3/m^3. The correlation for the red soil at bulk density of 1 600 kg/m^3 is similar with a R^2 of 0.93 and RMSE of 0.03 m^3/m^3. The R^2 and RMES for the densest black soil at bulk density of 1 600 kg/m^3 are 0.87 and 0.07 m^3/m^3 respectively. Considering the change insoil bulk density, the R^2 and RMSE in the models establishedfor the red soil are R^2=0.72 and RMSE=0.06 m^3/m^3,for the loam are R^2=0.70 and RMSE=0.07 m^3/m^3, and for the black soil are R^2=0.60 and RMSE =0.07 m^3/m^3. It is hence feasible to estimate soil surface water content using on digital camera.