Fig. 5From: Burn image segmentation based on Mask Regions with Convolutional Neural Network deep learning framework: more accurate and more convenientAtrous convolution. Atrous convolution with kernel size 3 × 3 and different rates. From left to right in the figure, the atrous rates are 1, 2, and 3. Standard convolution corresponds to atrous convolution with rate = 1. In this article, we set the rate as 2Back to article page