The topographic impact on land surface reflectance is an important factor influencing the practical use of remotely sensed data in forested areas. The sun-canopy-sensor (SCS) topographic correction model has been developed to reduce the topographic effect in forested regions, which is more appropriate than terrain-based methods because it considers the geotropic nature of tree growth. However, the SCS model, similar to the cosine correction, can not avoid an overcorrection problem. To address this problem, an empirical parameter C has been introduced into the SCS algorithm to form the SCS+C model. A well-known Minnaert constant might also play a role in reducing the overcorrection. In this paper, a new SCS+K topographic correction method by introducing the Minnaert constant k into the SCS model was proposed. The SCS+K model was applied to a TM forest image of a rugged site in Huailou County, Beijing, China. In comparison with the SCS and SCS+C algorithms, the SCS+K model performed markedly better than the SCS method and was comparable to the SCS+C approach. The results of this study showed that the SCS+K method can effectively reduce the topographic effect in forested terrain.