Skip to main content
. 2022 Aug 19;22(16):6226. doi: 10.3390/s22166226

Figure 2.

Figure 2

(a) The grid layout extracted by the repeated pattern detector and drawn on the original image. (b) The edges and defects are filled with grids. (c,d) The input labels and the one-hot encoded target labels of the semantic segmentation network.