Road Extraction by Deep Residual U-Net

Abstract

Road extraction from aerial images has been a hot research topic in the field of remote sensing image analysis. In this letter, a semantic segmentation neural network which combines the strengths of residual learning and U-Net is proposed for road area extraction. The network is built with residual units and has similar architecture to that of U-Net. The benefits of this model is two-fold: first, residual units ease training of deep networks. Second, the rich skip connections within the network could facilitate information propagation, allowing us to design networks with fewer parameters however better performance. We test our network on a public road dataset and compare it with U-Net and other two state of the art deep learning based road extraction methods. The proposed approach outperforms all the comparing methods, which demonstrates its superiority over recently developed state of the arts.

arXiv commentSubmitted to IEEE Geoscience and Remote Sensing Letters

References in corpus (1)

• Very Deep Convolutional Networks for Large-Scale Image Recognition

Cited by in corpus (150)

• ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data
• COVID-CT-Dataset: A CT Scan Dataset about COVID-19
• Recurrent Residual Convolutional Neural Network based on U-Net (R2U-Net) for Medical Image Segmentation
• Real-Time Polyp Detection, Localization and Segmentation in Colonoscopy Using Deep Learning
• Very high resolution canopy height maps from RGB imagery using self-supervised vision transformer and convolutional decoder trained on Aerial Lidar
• Video-SwinUNet: Spatio-temporal Deep Learning Framework for VFSS Instance Segmentation
• Fast Camouflaged Object Detection via Edge-based Reversible Re-calibration Network
• Topology optimization of 2D structures with nonlinearities using deep learning
• Landslide4Sense: Reference Benchmark Data and Deep Learning Models for Landslide Detection
• Road Segmentation in SAR Satellite Images with Deep Fully-Convolutional Neural Networks
• From W-Net to CDGAN: Bi-temporal Change Detection via Deep Learning Techniques
• FCN-Transformer Feature Fusion for Polyp Segmentation
• A multi-centre polyp detection and segmentation dataset for generalisability assessment
• Deep Learning Enabled Strain Mapping of Single-Atom Defects in 2D Transition Metal Dichalcogenides with Sub-picometer Precision
• CaraNet: Context Axial Reverse Attention Network for Segmentation of Small Medical Objects
• MIScnn: A Framework for Medical Image Segmentation with Convolutional Neural Networks and Deep Learning
• X-Net: Brain Stroke Lesion Segmentation Based on Depthwise Separable Convolution and Long-range Dependencies
• DDU-Net: Dual-Decoder-U-Net for Road Extraction Using High-Resolution Remote Sensing Images
• TransBTS: Multimodal Brain Tumor Segmentation Using Transformer
• PraNet: Parallel Reverse Attention Network for Polyp Segmentation
• Mapping horizontal and vertical urban densification in Denmark with Landsat time-series from 1985 to 2018: a semantic segmentation solution
• TransFuse: Fusing Transformers and CNNs for Medical Image Segmentation
• Optimized Deep Encoder-Decoder Methods for Crack Segmentation
• A Shallow U-Net Architecture for Reliably Predicting Blood Pressure (BP) from Photoplethysmogram (PPG) and Electrocardiogram (ECG) Signals
• DiResNet: Direction-aware Residual Network for Road Extraction in VHR Remote Sensing Images
• A Deep Neural Network for Pixel-Level Electromagnetic Particle Identification in the MicroBooNE Liquid Argon Time Projection Chamber
• ChemGrapher: Optical Graph Recognition of Chemical Compounds by Deep Learning
• BDG-Net: Boundary Distribution Guided Network for Accurate Polyp Segmentation
• DeepCMB: Lensing Reconstruction of the Cosmic Microwave Background with Deep Neural Networks
• Precipitation Nowcasting with Satellite Imagery
• City-Scale Road Extraction from Satellite Imagery v2: Road Speeds and Travel Times
• WeGleNet: A Weakly-Supervised Convolutional Neural Network for the Semantic Segmentation of Gleason Grades in Prostate Histology Images
• Energy-based Legged Robots Terrain Traversability Modeling via Deep Inverse Reinforcement Learning
• Automated crater detection with human level performance
• High-precision Density Mapping of Marine Debris and Floating Plastics via Satellite Imagery
• MimickNet, Matching Clinical Post-Processing Under Realistic Black-Box Constraints
• A global method to identify trees outside of closed-canopy forests with medium-resolution satellite imagery
• CS2-Net: Deep Learning Segmentation of Curvilinear Structures in Medical Imaging
• Road Extraction with Satellite Images and Partial Road Maps
• Deep Learning for Real-Time Aerodynamic Evaluations of Arbitrary Vehicle Shapes
• CASI: A Convolutional Neural Network Approach for Shell Identification
• Semantic Labeling of High Resolution Images Using EfficientUNets and Transformers
• ARPM-net: A novel CNN-based adversarial method with Markov Random Field enhancement for prostate and organs at risk segmentation in pelvic CT images
• A modular U-Net for automated segmentation of X-ray tomography images in composite materials
• Edge-aware Feature Aggregation Network for Polyp Segmentation
• Software Engineering Practice in the Development of Deep Learning Applications
• Deep-NFA: a Deep $\text{a contrario}$ Framework for Small Object Detection
• Improving Across-Dataset Brain Tissue Segmentation Using Transformer
• A Deep Learning-Based Fully Automated Pipeline for Regurgitant Mitral Valve Anatomy Analysis From 3D Echocardiography
• Mass Estimation of Galaxy Clusters with Deep Learning II: CMB Cluster Lensing
• SU-Net: Pose estimation network for non-cooperative spacecraft on-orbit
• Reconstructing Patchy Reionization with Deep Learning
• HSMix: Hard and Soft Mixing Data Augmentation for Medical Image Segmentation
• Transformers-based architectures for stroke segmentation: A review
• DFENet: A Novel Dimension Fusion Edge Guided Network for Brain MRI Segmentation
• Mapping Urban Villages in China: Progress and Challenges
• CFPNet-M: A Light-Weight Encoder-Decoder Based Network for Multimodal Biomedical Image Real-Time Segmentation
• ROSE: A Retinal OCT-Angiography Vessel Segmentation Dataset and New Model
• OCTAve: 2D en face Optical Coherence Tomography Angiography Vessel Segmentation in Weakly-Supervised Learning with Locality Augmentation
• EigenRank by Committee: A Data Subset Selection and Failure Prediction paradigm for Robust Deep Learning based Medical Image Segmentation
• Generalizable synthetic MRI with physics-informed convolutional networks
• Reconstructing Cosmic Polarization Rotation with ResUNet-CMB
• Inner Cell Mass and Trophectoderm Segmentation in Human Blastocyst Images using Deep Neural Network
• Self-supervised Remote Sensing Images Change Detection at Pixel-level
• UTRNet: High-Resolution Urdu Text Recognition In Printed Documents
• Automatic Detection of Interplanetary Coronal Mass Ejections in Solar Wind In Situ Data
• High-quality Speech Synthesis Using Super-resolution Mel-Spectrogram
• NL-LinkNet: Toward Lighter but More Accurate Road Extraction with Non-Local Operations
• Holistic Grid Fusion Based Stop Line Estimation
• Shallow Attention Network for Polyp Segmentation
• A Comprehensive Study on Colorectal Polyp Segmentation with ResUNet++, Conditional Random Field and Test-Time Augmentation
• Productive Crop Field Detection: A New Dataset and Deep Learning Benchmark Results
• Amplitude SAR Imagery Splicing Localization
• Real-time guidewire tracking and segmentation in intraoperative x-ray
• City-scale Road Extraction from Satellite Imagery
• Semantic Labeling of Large-Area Geographic Regions Using Multi-View and Multi-Date Satellite Images and Noisy OSM Training Labels
• Scribble-based Weakly Supervised Deep Learning for Road Surface Extraction from Remote Sensing Images
• Deep Convolutional Neural Networks with Spatial Regularization, Volume and Star-shape Priori for Image Segmentation
• Image Segmentation of Zona-Ablated Human Blastocysts
• Sat2Graph: Road Graph Extraction through Graph-Tensor Encoding
• Extreme Augmentation : Can deep learning based medical image segmentation be trained using a single manually delineated scan?
• CoKe: Localized Contrastive Learning for Robust Keypoint Detection
• Volumetric Lung Nodule Segmentation using Adaptive ROI with Multi-View Residual Learning
• DanceCam: atmospheric turbulence mitigation in wide-field astronomical images with short-exposure video streams
• Deep Fusion of Local and Non-Local Features for Precision Landslide Recognition
• Lensing reconstruction from the cosmic microwave background polarization with machine learning
• Road detection via a dual-task network based on cross-layer graph fusion modules
• A fully 3D multi-path convolutional neural network with feature fusion and feature weighting for automatic lesion identification in brain MRI images
• Frontiers in Intelligent Colonoscopy
• Learning Context-Based Non-local Entropy Modeling for Image Compression
• Adaptively Enhancing Facial Expression Crucial Regions via Local Non-Local Joint Network
• AG-CUResNeSt: A Novel Method for Colon Polyp Segmentation
• Deep learning modelling of manufacturing and build variations on multi-stage axial compressors aerodynamics
• Dual Encoder Fusion U-Net (DEFU-Net) for Cross-manufacturer Chest X-ray Segmentation
• No Surprises: Training Robust Lung Nodule Detection for Low-Dose CT Scans by Augmenting with Adversarial Attacks
• PVTAdpNet: Polyp Segmentation using Pyramid vision transformer with a novel Adapter block
• Automated Inline Analysis of Myocardial Perfusion MRI with Deep Learning
• Automated quasar continuum estimation using neural networks: a comparative study of deep-learning architectures
• DualSwinUnet++: An Enhanced Swin-Unet Architecture With Dual Decoders For PTMC Segmentation
• Does pre-training on brain-related tasks results in better deep-learning-based brain age biomarkers?
• Self-supervised component separation for the extragalactic submillimeter sky
• Deep Learning Seismic Substructure Detection using the Frozen Gaussian Approximation
• Few-Shot Defect Segmentation Leveraging Abundant Normal Training Samples Through Normal Background Regularization and Crop-and-Paste Operation
• CXR-Net: An Artificial Intelligence Pipeline for Quick Covid-19 Screening of Chest X-Rays
• Beacon2Science: Enhancing STEREO/HI beacon data with machine learning for efficient CME tracking
• Lung Segmentation in Chest X-rays with Res-CR-Net
• Very High Resolution Land Cover Mapping of Urban Areas at Global Scale with Convolutional Neural Networks
• Transport-Related Surface Detection with Machine Learning: Analyzing Temporal Trends in Madrid and Vienna
• BeSt-LeS: Benchmarking Stroke Lesion Segmentation using Deep Supervision
• Semi-supervised Semantic Segmentation of Prostate and Organs-at-Risk on 3D Pelvic CT Images
• Aerial Images Meet Crowdsourced Trajectories: A New Approach to Robust Road Extraction
• Coronary Wall Segmentation in CCTA Scans via a Hybrid Net with Contours Regularization
• Detecting anthropogenic cloud perturbations with deep learning
• SANet:Superpixel Attention Network for Skin Lesion Attributes Detection
• Landmark detection in Cardiac Magnetic Resonance Imaging Using A Convolutional Neural Network
• Pix2Streams: Dynamic Hydrology Maps from Satellite-LiDAR Fusion
• Acoustic Echo Cancellation using Residual U-Nets
• AI Powered Road Network Prediction with Multi-Modal Data
• Fitzpatrick Thresholding for Skin Image Segmentation
• ForCM: Forest Cover Mapping from Multispectral Sentinel-2 Image by Integrating Deep Learning with Object-Based Image Analysis
• OMSN and FAROS: OCTA Microstructure Segmentation Network and Fully Annotated Retinal OCTA Segmentation Dataset
• Wound Healing Modeling Using Partial Differential Equation and Deep Learning
• Towards Robotic Knee Arthroscopy: Multi-Scale Network for Tissue-Tool Segmentation
• Self-supervised SAR-optical Data Fusion and Land-cover Mapping using Sentinel-1/-2 Images
• Improving Lesion Detection by exploring bias on Skin Lesion dataset
• A novel Deep Structure U-Net for Sea-Land Segmentation in Remote Sensing Images
• Predicting Clinical Outcomes in COVID-19 using Radiomics and Deep Learning on Chest Radiographs: A Multi-Institutional Study
• Shape related constraints aware generation of Mechanical Designs through Deep Convolutional GAN
• Generation and Simulation of Yeast Microscopy Imagery with Deep Learning
• Extreme Rotation Estimation using Dense Correlation Volumes
• Remote Estimation of Free-Flow Speeds
• Improving segmentation of calcified and non-calcified plaques on CCTA-CPR scans via masking of the artery wall
• Innovative Quantitative Analysis for Disease Progression Assessment in Familial Cerebral Cavernous Malformations
• Full-scale Representation Guided Network for Retinal Vessel Segmentation
• Joint multiband deconvolution for Euclid and Vera C. Rubin images
• Fracture interactive geodesic active contours for bone segmentation
• Learning to Predict Contact Force Distributions from Vision Leveraging Object Geometry Priors
• Attention Toward Neighbors: A Context Aware Framework for High Resolution Image Segmentation
• AOSLO-net: A deep learning-based method for automatic segmentation of retinal microaneurysms from adaptive optics scanning laser ophthalmoscope images
• A new smart-cropping pipeline for prostate segmentation using deep learning networks
• Topo2vec: Topography Embedding Using the Fractal Effect
• Appearance Editing with Free-viewpoint Neural Rendering
• A grid-point detection method based on U-net for a structured light system
• CleftNet: Augmented Deep Learning for Synaptic Cleft Detection from Brain Electron Microscopy
• Crowd-Sourced Road Quality Mapping in the Developing World
• An Uncertainty-aware Hierarchical Probabilistic Network for Early Prediction, Quantification and Segmentation of Pulmonary Tumour Growth
• Seeing as Humans Do: Learning from Motion to Segment Anything Without Supervision
• Rethinking Radiology: An Analysis of Different Approaches to BraTS
• Road Mapping in Low Data Environments with OpenStreetMap
• Trident Segmentation CNN: A Spatiotemporal Transformation CNN for Punctate White Matter Lesions Segmentation in Preterm Neonates