The survey and evaluation of cultivated land quality is an important basis for the protection of the “three in one” of the quantity, quality and ecology of cultivated land, and also an important cornerstone for ensuring national food security and maintaining sustainable agricultural development. In the mountainous and hilly areas, the estimation of the attribute values of the grading factors in the survey and evaluation of cultivated land quality grades is affected by the barriers of large and mediumsized mountain bodies, resulting in the misalignment of the attribute assignment results of the grading elements. Taking the Pingan District of Qinghai Province as the research area, the method of estimating the attribute value of the grading factor based on the barrier factor was studied. The crossvalidation was performed with the inverse distance weight interpolation and the spline function interpolation method. The results showed that different interpolation methods were basically consistent with the simulation of the spatial distribution trend of three grading factors. Because of the difference between the characteristics of the original data and the interpolation principle, the interpolation method based on the barrier factors was inconsistent to different soil properties, and the most suitable interpolation method for soil organic matter and effective soil thickness was the spline function interpolation based on the barrier factor. In the soil pH value interpolation, the inverse distance weight interpolation accuracy based on the barrier factor was the highest; the interpolation accuracy of soil organic matter and effective soil thickness in the study area was higher than the average absolute error, the average relative error, the root mean square error, and the relative root mean square error of the spline function interpolation without considering the blocking factor, respectively, by 11.23%, 10.98%, 7.54%, 9.20%, and 15.08%, 11.74%, 17.41%, 9.40%. Interpolation of organic matter and effective soil layer was performed by interpolation of spline function based on barrier factors. The determination coefficients between the measured and predicted values were 0.927 and 0.901, respectively.