This repository got inspiration and insight on repository deep_residual_unet and paper Efficient Multi-Class Semantic Segmentation of High Resolution Aerial Imagery with Dilated LinkNet. The model mostly the same as Deep Residual U-Net, but on the bottleneck we added Dilated Block from Dilated LinkNet. The usage of Residual U-Net are for minimize the information loss on segmentation and Dilated Block are for bigger receptive field with minimum weights. For more detail on both architectures please check both papers Road Extraction by Deep Residual U-Net and Efficient Multi-Class Semantic Segmentation of High Resolution Aerial Imagery with Dilated LinkNet.