Road networks as collections of minimum cost paths

Abstract We present a probabilistic representation of network structures in images. Our target application is the extraction of urban roads from aerial images. Roads appear as thin, elongated, partially curved structures forming a loopy graph, and this complex layout requires a prior that goes beyond standard smoothness and co-occurrence assumptions. In the proposed model the network is represented as a union of 1D paths connecting distant (super-)pixels. A large set of putative candidate paths is constructed in such a way that they include the true network as much as possible, by searching for minimum cost paths in the foreground (road) likelihood. Selecting the optimal subset of candidate paths is posed as MAP inference in a higher-order conditional random field. Each path forms a higher-order clique with a type of clique potential, which attracts the member nodes of cliques with high cumulative road evidence to the foreground label. That formulation induces a robust P N -Potts model, for which a global MAP solution can be found efficiently with graph cuts. Experiments with two road data sets show that the proposed model significantly improves per-pixel accuracies as well as the overall topological network quality with respect to several baselines.

[1]  Paria Mehrani,et al.  Superpixels and Supervoxels in an Energy Optimization Framework , 2010, ECCV.

[2]  Pushmeet Kohli,et al.  Graph Cut Based Inference with Co-occurrence Statistics , 2010, ECCV.

[3]  Eugene W. Myers,et al.  Automated Reconstruction of Neuronal Morphology Based on Local Geometrical and Global Structural Models , 2011, Neuroinformatics.

[4]  Sanja Fidler,et al.  Describing the scene as a whole: Joint object detection, scene classification and semantic segmentation , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[5]  A. Gruen,et al.  Semi-Automatic Linear Feature Extraction by Dynamic Programming and LSB-Snakes , 1997 .

[6]  George Vosselman,et al.  Road tracing by profile matching and Kaiman filtering , 1995 .

[7]  Christian Heipke,et al.  Graph-supported verification of road databases , 2004 .

[8]  Pascal Fua,et al.  Reconstructing Loopy Curvilinear Structures Using Integer Programming , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Chunsun Zhang.  Towards an operational system for automated updating of road databases by integration of imagery and geodata , 2004 .

[10]  Geoffrey E. Hinton,et al.  Learning to Label Aerial Images from Noisy Data , 2012, ICML.

[11]  Peter Wonka,et al.  Road Network Extraction and Intersection Detection From Aerial Images by Tracking Road Footprints , 2007, IEEE Transactions on Geoscience and Remote Sensing.

[12]  Sahar Movaghati,et al.  Road Extraction From Satellite Images Using Particle Filtering and Extended Kalman Filtering , 2010, IEEE Transactions on Geoscience and Remote Sensing.

[13]  Pushmeet Kohli,et al.  Robust Higher Order Potentials for Enforcing Label Consistency , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[14]  Peter Doucette,et al.  Automated Road Extraction from High Resolution Multispectral Imagery , 2004 .

[15]  Pascal Fua,et al.  SLIC Superpixels Compared to State-of-the-Art Superpixel Methods , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Aluir Porfírio Dal Poz,et al.  Object-Space Road Extraction in Rural Areas Using Stereoscopic Aerial Images , 2012, IEEE Geoscience and Remote Sensing Letters.

[17]  Josiane Zerubia,et al.  Point processes for unsupervised line network extraction in remote sensing , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Martin A. Fischler,et al.  Detection of roads and linear structures in low-resolution aerial imagery using a multisource knowledge integration technique☆ , 1981 .

[19]  Edsger W. Dijkstra,et al.  A note on two problems in connexion with graphs , 1959, Numerische Mathematik.

[20]  Christian Heipke,et al.  EVALUATION OF AUTOMATIC ROAD EXTRACTION , 2007 .

[21]  Jan Dirk Wegner,et al.  A Higher-Order CRF Model for Road Network Extraction , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[22]  Vladimir Kolmogorov,et al.  Graph cut based image segmentation with connectivity priors , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[23]  Edward M. Mikhail,et al.  Extracting Urban Road Networks from High-resolution True Orthoimage and Lidar , 2008 .

[24]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[25]  Pushmeet Kohli,et al.  Minimizing sparse higher order energy functions of discrete variables , 2009, CVPR.

[26]  Antonio Criminisi,et al.  Object categorization by learned universal visual dictionary , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[27]  Aluir Porfírio Dal Poz,et al.  Three-Dimensional Semiautomatic Road Extraction From a High-Resolution Aerial Image by Dynamic-Programming Optimization in the Object Space , 2010, IEEE Geoscience and Remote Sensing Letters.

[28]  C. Heipke,et al.  Semi-automatic quality control of topographic data sets , 2012 .

[29]  Florent Lafarge,et al.  Geometric Feature Extraction by a Multimarked Point Process , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[30]  Stefan Hinz,et al.  Automatic extraction of urban road networks from multi-view aerial imagery , 2003 .

[31]  Gertjan J. Burghouts,et al.  Material-specific adaptation of color invariant features , 2009, Pattern Recognit. Lett..

[32]  Christian Heipke,et al.  Network snakes: graph-based object delineation with active contour models , 2010, Machine Vision and Applications.

[33]  Ivan Laptev,et al.  Automatic extraction of roads from aerial images based on scale space and snakes , 2000 .

[34]  Badrinath Roysam,et al.  Novel 4-D Open-Curve Active Contour and curve completion approach for automated tree structure extraction , 2011, CVPR 2011.

[35]  Pascal Fua,et al.  Automated Reconstruction of Dendritic and Axonal Trees by Global Optimization with Geometric Priors , 2011, Neuroinformatics.

[36]  Haihong Li,et al.  Road extraction from aerial and satellite images by dynamic programming , 1995 .

[37]  Suya You,et al.  Delineation and geometric modeling of road networks , 2010 .

[38]  Marie-Pierre Jolly,et al.  Interactive graph cuts for optimal boundary & region segmentation of objects in N-D images , 2001, Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001.

[39]  Jan Dirk Wegner,et al.  Mind the Gap: Modeling Local and Global Context in (Road) Networks , 2014, GCPR.

[40]  Deniz Erdogmus,et al.  Principal Curves as Skeletons of Tubular Objects , 2011, Neuroinformatics.

[41]  Anthony J. Yezzi,et al.  Vessels as 4-D Curves: Global Minimal 4-D Paths to Extract 3-D Tubular Surfaces and Centerlines , 2007, IEEE Transactions on Medical Imaging.

[42]  José A. Malpica,et al.  An automatic method for road extraction in rural and semi-urban areas starting from high resolution satellite imagery , 2005, Pattern Recognit. Lett..

[43]  Ruzena Bajcsy,et al.  Computer Recognition of Roads from Satellite Pictures , 1976, IEEE Transactions on Systems, Man, and Cybernetics.

[44]  Wenzhong Shi,et al.  Road Centerline Extraction From High-Resolution Imagery Based on Shape Features and Multivariate Adaptive Regression Splines , 2013, IEEE Geoscience and Remote Sensing Letters.

[45]  Pascal Fua,et al.  Automated reconstruction of tree structures using path classifiers and Mixed Integer Programming , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[46]  Florent Lafarge,et al.  Recovering Line-Networks in Images by Junction-Point Processes , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[47]  Christian Heipke,et al.  Road Network Extraction in Suburban Areas , 2012 .

[48]  A. Baumgartner.  EFFICIENT METHODS AND INTERFACES FOR ROAD TRACKING , 2002 .

[49]  Josiane Zerubia,et al.  A Gibbs Point Process for Road Extraction from Remotely Sensed Images , 2004, International Journal of Computer Vision.

[50]  Laurent D. Cohen,et al.  Tubular Structure Segmentation Based on Minimal Path Method and Anisotropic Enhancement , 2011, International Journal of Computer Vision.

[51]  Konrad Schindler,et al.  SEMANTIC SEGMENTATION OF AERIAL IMAGES IN URBAN AREAS WITH CLASS-SPECIFIC HIGHER-ORDER CLIQUES , 2015 .

[52]  Geoffrey E. Hinton,et al.  Learning to Detect Roads in High-Resolution Aerial Images , 2010, ECCV.

[53]  Cem Ünsalan,et al.  Road Network Detection Using Probabilistic and Graph Theoretical Methods , 2012, IEEE Transactions on Geoscience and Remote Sensing.