Posterior cruciate ligament (PCL) is important for the knee stability. The problem seems to lie in the fact that it is difficult to segment the PCL from a magnetic resonance (MR) image. To reduce massive measurement time and subjective judgment error, this paper proposes a segmentation method of the PCL in knee joint from MR image and computed tomography (CT) image using anatomical information. In the experiment, a normal knee was analyzed by the proposed method. In the result, the segmented PCL was visually observed. We concluded that the proposed method can segment the PCL from the knee MR image.