Abstract: Two methods for calculating similarity for spatial directions between areal objects in raster data are given: one is based on the features of raster data and the directional matrix to calculate the similarity for spatial directions; the other is based on the variation of angle between each raster cell and a reference object to calculate the similarity for spatial directions. The two methods simplify the calculation of similarity given by Goyal for spatial directions based on direction matrix model, or overcome the limitation of direction small change. The two methods have broader applicability to calculating the similarity for spatial directions.