Fig. 4
From: Deep learning approach for automatic segmentation of ulna and radius in dual-energy X-ray imaging
Visual comparison of the ulna and radius segmentation results using different methods on the testing set. Columns from left to right: input image, ground truth, U-Net, FCN, and proposed method. The first and second rows show the low-energy X-ray images, and the third and fourth rows show the high-energy X-ray images. The red circle denotes the region of segmentation error