Skip to main content
. 2021 Aug 13;10(11):641–661. doi: 10.1089/wound.2018.0937

Figure 10.

Figure 10.

(from left to right) The cropped image is taken as input by the neural network- At the output pixel-wise probabilities of wound segment are provided in gray levels the and final mask is obtained by setting a threshold of 0.5 on every pixel, to compare with the ground truth mask displayed in the last column. Wang 2015.62