In addressing the prevalent issue of terrain occlusion in satellite oblique images, this paper proposes a high-resolution occlusion detection method based on the Rational Polynomial Coefficients (RPC) model. The proposed approach utilizes an irregular triangular mesh to represent the three-dimensional terrain surface and employs ray tracing techniques to accurately detect the intersections of incident rays with this mesh. This process facilitates the determination of occlusion relationships among the intersection points. To enhance the efficiency of occlusion detection, a strategy is developed that minimizes the search area within the terrain. Experimental validation is performed using both real oblique satellite images and simulated satellite images. The results demonstrate that the proposed method achieves an occlusion detection accuracy exceeding 97%, the area affected by terrain occluded area can be identified effectively and the mask image of terrain occluded area can be generated automatically.